Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Wed, 29 Aug 2012 09:35:36 -0400 (EDT), Stan Hoeppner wrote: Stephen, most PC IO plates are removable, have been for well over 15 years. Once you remove the motherboard you simply stick your thumb on the interior side of the IO panel and pop it out the back. New mobos come with their own panel. They just pop in. They friction fit, having little nubs the size of a pin head, usually 2 on each side. You install it gently, pushing only on the outside edges until all the nubs have snapped in. Push in the middle and you may destroy it. they're pretty flimsy. That's good to know. Thanks. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/869328987.788493.1346446652586.javamail.r...@md01.wow.synacor.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/28/2012 2:01 PM, Martin Steigerwald wrote: Hi Stan, Am Montag, 27. August 2012 schrieb Stan Hoeppner: For a desktop user workload, there will be no noticeable performance difference, because such applications don't do parallel IO. Are you sure about 1) desktop applications I/O behavior and 2) NCQ? 1) I see noticeable difference for my Intel SSD 320 with different iodepth values in fio job. Shall we now split hairs, and have a 500 post discussion WRT the definition of desktop user workload? If you consider fio, iozone, bonnie++ etc to be part of a typical desktop user workload then you're simply out of touch with the real world Martin. ;) Do you have a father/mother brother/sister wife/children? Do any of them run fio on a regular basis? These people represent the typical desktop user workload. Sysadmins, devs, power users, do not. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/503e17f6.7040...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/28/2012 3:11 PM, Stephen Powell wrote: On Sun, 26 Aug 2012 09:48:27 -0400 (EDT), Stephen Powell wrote: What I meant was that I may buy the new mobo, processor, and RAM that you suggested and put it in this second machine I'm talking about. Well, I've looked into it, and that's not going to work. The mobo you suggested has stuff on it for which there are no connector holes on the back of the case. Stephen, most PC IO plates are removable, have been for well over 15 years. Once you remove the motherboard you simply stick your thumb on the interior side of the IO panel and pop it out the back. New mobos come with their own panel. They just pop in. They friction fit, having little nubs the size of a pin head, usually 2 on each side. You install it gently, pushing only on the outside edges until all the nubs have snapped in. Push in the middle and you may destroy it. they're pretty flimsy. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/503e1aa8.9010...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
Am Mittwoch, 29. August 2012 schrieb Stan Hoeppner: On 8/28/2012 2:01 PM, Martin Steigerwald wrote: Hi Stan, Am Montag, 27. August 2012 schrieb Stan Hoeppner: For a desktop user workload, there will be no noticeable performance difference, because such applications don't do parallel IO. Are you sure about 1) desktop applications I/O behavior and 2) NCQ? 1) I see noticeable difference for my Intel SSD 320 with different iodepth values in fio job. Shall we now split hairs, and have a 500 post discussion WRT the definition of desktop user workload? If you consider fio, iozone, bonnie++ etc to be part of a typical desktop user workload then you're simply out of touch with the real world Martin. ;) Do you have a father/mother brother/sister wife/children? Do any of them run fio on a regular basis? These people represent the typical desktop user workload. Sysadmins, devs, power users, do not. While a typical desktop workload may not induce anything near an IO depth of 64, I still think several desktop applications accessing the disk at once can have the Linux kernel block layer use an iodepth 1 on the disk. So maybe testing with I/O depth upto 10 or so make sense even for desktop workloads. -- Martin 'Helios' Steigerwald - http://www.Lichtvoll.de GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7 -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/201208291607.06354.mar...@lichtvoll.de
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/27/2012 6:12 PM, Stephen Powell wrote: On Mon, 27 Aug 2012 01:42:54 -0400 (EDT), Stan Hoeppner wrote: Stephen Powell wrote: I know enough to be dangerous, but not always enough to be competent. That's why I opened this thread in the first place. In this case the solution is as simple as downloading and reading the manual for your board: http://ph.academicdirect.ro/ISB_SE7500CW2.pdf The memory specs are on page 17. Stan, you are a wealth of information. I didn't search for the motherboard manual because, for some reason, I didn't think they were published online. Maybe I did an extensive search in the past for a motherboard or chipset manual and found nothing; so I already had it in my head that such things weren't available. But in this case I am happy to be proved wrong. It's quite common for someone around the globe to re-host the manual for an EOL/obsolete product. Sometimes they're easy to find, sometimes not so easy. Your manual was pretty easy with a basic Google search. I found it not long after this thread began. By the way, there's something I don't understand. A 32-bit processor can only access 4G of real (extended) memory, right? So why are there motherboards available for 32-bit processors that support installing more than 4G of RAM? What good is memory that the processor can't address? The processor can address up to 64GB using PAE: http://en.wikipedia.org/wiki/Physical_Address_Extension PAE, BTW has been with us in every Intel and AMD CPU since the Pentium Pro and K6, since 1996, 16 years, one year after I started working on computers professionally. All new AMD/Intel x86 CPUs for many years have had the x86-64 ISA (instruction set architecture), so PAE is rarely used today. Before AMD introduced the the x86-64 ISA with the Opteron, if one wanted large memory in a server PAE was the only option in the x86 world. Only a few systems took serious advantage of PAE, almost all for database work. The Data General Aviion 25000 sported up to 32 Pentium Pro CPUs and 64GB RAM, running their proprietary DG/UX SVR4 Unix variant; exclusively used in the health care industry for patient record databases, etc. Axil (acquired by HP) had an 8-way Pentium Pro box with up to 8 or 16GB RAM which they sold as an MS SQL server solution. Unisys had a series of ES7000 machines with up to 32 Xeon processors and 64GB of RAM, also sold as an MS SQL server platform though Linux was supported. These were really neat machines. Somewhat off topic, but very interesting, these 32-way Unisys machines were quite unique at the time as they could be partitioned into independent physical servers (not virtual) of 4, 8, 12, 16, 20, 24, 28, or 32 CPUs each. This was possible because the system was built around 4 CPU cellular multiprocessing boards interconnected by a custom cross bar switch network which was connected to a single high bandwidth flat memory subsystem composed of multiple interleaved channels. It's memory bandwidth of 20GB/s was many times higher than any x86 server at that time as they all used a single P6 bus, with only 1GB/s bandwidth. 20GB/s is peanuts today given just two channels of DDR3-1333 have just over 20GB/s, but back then, in the late 1990s, this was huge. This CMP design also allowed assigning different amounts of memory to each of the hosts, with the firmware and custom crossbar chipset setting up the fences in the physical memory map. Individual PCI buses and IO devices could be assigned to any of the partitioned servers. In the first models, a console module had to be installed which included VGA, KB, mouse ports and each was controlled via a KVM switch. Later models had a much more intelligent solution in the form of a single system controller. This also facilitated the ability to cluster multiple sets of two physical hosts (up to 4 clusters per server) within a single server using system memory as the cluster interconnect, with latency thousands of times lower and bandwidth thousands of times higher than the fastest network interconnects at that time, this became immensely popular with Unisys customers, many running multiple clustered MS SQL servers within one ES7000 mainframe. Note I used the word mainframe. All of this unique technology and scalability Unisys brought to x86, MS Windows and, to a lesser degree, Linux, was directly derived from Unisys' ClearPath mainframe technology, where such features had been common for years, as well as on IBM, Fujitsu, and other mainframes. Sadly, due to market realities and diminished customer demand for large monolithic servers, the biggest x86 box Unisys now sells is an 8-way 4U Xeon box. Though with up to 80 cores, 332x times the memory bandwidth, and similarly higher IO bus bandwidth, it runs circles around the monster 32 socket mainframe style boxes of yesterday. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Tue, 28 Aug 2012, Stan Hoeppner wrote: Sadly, due to market realities and diminished customer demand for large monolithic servers, the biggest x86 box Unisys now sells is an 8-way 4U Xeon box. Though with up to 80 cores, 332x times the memory bandwidth, and similarly higher IO bus bandwidth, it runs circles around the monster 32 socket mainframe style boxes of yesterday. You can go to SGI for suitably large n-way Linux servers. An SGI UV 2k rack will go up to 256 Intel E5 processors (2048 cores, 4096 threads), 64TB RAM. http://www.sgi.com/products/servers/uv/ I don't think the UV has mainframe-style hardware partitioning. It is a big NUMA box with a complex hypercube-like node topology. -- One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie. -- The Silicon Valley Tarot Henrique Holschuh -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120828104000.ga...@khazad-dum.debian.net
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Tue, 28 Aug 2012 04:04:20 -0400 (EDT), Stan Hoeppner wrote: Stephen Powell wrote: By the way, there's something I don't understand. A 32-bit processor can only access 4G of real (extended) memory, right? So why are there motherboards available for 32-bit processors that support installing more than 4G of RAM? What good is memory that the processor can't address? The processor can address up to 64GB using PAE: http://en.wikipedia.org/wiki/Physical_Address_Extension That explains it. Thanks. Thanks also to the other members of the list who responded regarding PAE. ... Note I used the word mainframe. All of this unique technology and scalability Unisys brought to x86, MS Windows and, to a lesser degree, Linux, was directly derived from Unisys' ClearPath mainframe technology, where such features had been common for years, as well as on IBM, Fujitsu, and other mainframes. ... I work on IBM mainframes as my day job. 32-bit IBM mainframes have a 2 GB address limit. (That's because, although the registers for manipulating binary integer numbers are 32-bits wide, the address bus has only 31 address lines.) Physical memory in excess of 2 GB could be installed though. You could use the extra memory in two ways. (1) IBM provides virtual-machine-level virtualization built-in to the hardware microcode, known as PR/SM (Processor Resource / System Manager). This can create what are known as LPARs (Logical PARtitions). You can divide up the physical memory between LPARs, so that each LPAR does not exceed 2G. (2) You can configure some of the memory, either in basic mode or in LPAR mode, as expanded memory. (Think back to the expanded memory boards in the days of DOS.) This in known in IBM lingo as XSTORE memory as opposed to CSTORE memory. Some IBM mainframe operating systems, such as VM, can use XSTORE memory as a high-speed paging device. Then, some applications, such as DB2, can use hyperspaces or VM data spaces as ways to exploit memory above 2 GB. I was aware of these techniques in the mainframe world to exploit memory above 2 GB, but I'm much less familiar with the world of PC CPUs. BTW, IBM's 64-bit machines no longer offer a basic mode. LPAR mode is required on all 64-bit machines, AFAIK. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1943825377.705792.1346163449270.javamail.r...@md01.wow.synacor.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
shawn wilson ag4ve...@gmail.com writes: On Tue, Aug 28, 2012 at 1:34 AM, Bob Proulx b...@proulx.com wrote: Stephen Powell wrote: By the way, there's something I don't understand. A 32-bit processor can only access 4G of real (extended) memory, right? So why are there motherboards available for 32-bit processors that support installing more than 4G of RAM? What good is memory that the processor can't address? With PAE (physical address extensions) the processor *can* address more than 4G of ram. A single process is still limited to 32-bits which usually works out effectively to 3G of ram but the operating system can make use of more than this. It can be used for filesystem buffer cache and for multiple 3G programs. A machine with 6G of ram for example could run two 3G program at the same time and hold them both in memory without swapping. Or run one 3G program and still have 3G for the system to use in filesystem buffer cache. With PAE having more than 4G of memory is quite useful. Using PAE does have a small performance impact. It slows things down by 2%-3% in my use cases. But the increase in ram for buffers usually more than makes up for the differences. iirc, pae is only 48 bits too. only meaning 256 terabytes in this case... I'll be very surprised to ever see a 32 bit processor that can make effective use of that much memory. -- Erwin, have you seen the cat? -- Mrs. Shroedinger -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1bligzghqq@pfeifferfamily.net
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/28/2012 5:40 AM, Henrique de Moraes Holschuh wrote: On Tue, 28 Aug 2012, Stan Hoeppner wrote: Sadly, due to market realities and diminished customer demand for large monolithic servers, the biggest x86 box Unisys now sells is an 8-way 4U Xeon box. Though with up to 80 cores, 332x times the memory bandwidth, and similarly higher IO bus bandwidth, it runs circles around the monster 32 socket mainframe style boxes of yesterday. You can go to SGI for suitably large n-way Linux servers. An SGI UV 2k rack will go up to 256 Intel E5 processors (2048 cores, 4096 threads), 64TB RAM. http://www.sgi.com/products/servers/uv/ Do I hear an echo? ;) I mention the UV and Infinite Storage products here quite often. The UV and its ancestors, the Altix 3xxx/4xxx IA64 beasts, are a different class of machine sold into a different market. The SGI boxen are used exclusively by HPC customers. The ES7000 and Aviion 25K were used exclusively in traditional commercial business environments, nearly exclusively hosting databases, although I did hear of a few large MS Exchange deployments on 16-way ES7000 boxen. I don't think the UV has mainframe-style hardware partitioning. It is a big NUMA box with a complex hypercube-like node topology. The CC-NUMA members of the Altix line can all be partitioned into nodes via programming the NUMALink routers. Once partitioned, an application can still span all processors, communicating via MPI and shmem over the ultra fast NUMALink interconnect. The MIPS/Irix predecessors, the Origin 2000/3000 could also be partitioned this way, as well as the IA64 Altix 3xxx/4xxx. It is relatively common for customers with very large SGI NUMA machines to partition them from day one, so a failure on a single NUMA node doesn't take down the entire system. This isn't quite the same as traditional mainframe partitioning, as you can't independently assign memory and IO buses to different nodes the way you can with many mainframe systems. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/503cfc08.3090...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
Joe Pfeiffer pfeif...@cs.nmsu.edu writes: shawn wilson ag4ve...@gmail.com writes: On Tue, Aug 28, 2012 at 1:34 AM, Bob Proulx b...@proulx.com wrote: Stephen Powell wrote: By the way, there's something I don't understand. A 32-bit processor can only access 4G of real (extended) memory, right? So why are there motherboards available for 32-bit processors that support installing more than 4G of RAM? What good is memory that the processor can't address? With PAE (physical address extensions) the processor *can* address more than 4G of ram. A single process is still limited to 32-bits which usually works out effectively to 3G of ram but the operating system can make use of more than this. It can be used for filesystem buffer cache and for multiple 3G programs. A machine with 6G of ram for example could run two 3G program at the same time and hold them both in memory without swapping. Or run one 3G program and still have 3G for the system to use in filesystem buffer cache. With PAE having more than 4G of memory is quite useful. Using PAE does have a small performance impact. It slows things down by 2%-3% in my use cases. But the increase in ram for buffers usually more than makes up for the differences. iirc, pae is only 48 bits too. only meaning 256 terabytes in this case... I'll be very surprised to ever see a 32 bit processor that can make effective use of that much memory. Sorry to follow up my own post -- I hit send too fast. My response was given the assumption that PAE gave a 48 bit physical address space; of course it doesn't. It gives 36 bits -- 64 GB, which isn't nearly so outlandish. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1b7gsjndia@pfeifferfamily.net
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Tue, Aug 28, 2012 at 4:04 AM, Stan Hoeppner s...@hardwarefreak.com wrote: It's memory bandwidth of 20GB/s was many times higher than any x86 server at that time as they all used a single P6 bus, with only 1GB/s bandwidth. 20GB/s is peanuts today given just two channels of DDR3-1333 have just over 20GB/s, but back then, in the late 1990s, this was huge. This CMP design also allowed assigning different amounts of memory to each of the hosts, with the firmware and custom crossbar chipset setting up the fences in the physical memory map. Individual PCI buses and IO devices could be assigned to any of the partitioned servers. In the first models, a console module had to be installed which included VGA, KB, mouse ports and each was controlled via a KVM switch. Later models had a much more intelligent solution in the form of a single system controller. This also facilitated the ability to cluster multiple sets of two physical hosts (up to 4 clusters per server) within a single server using system memory as the cluster interconnect, with latency thousands of times lower and bandwidth thousands of times higher than the fastest network interconnects at that time, this became immensely popular with Unisys customers, many running multiple clustered MS SQL servers within one ES7000 mainframe. wasn't the 20GB/s infiniband introduced in the late 90s / early 2k? that should about measure up to what you're describing, but with tons more scalability, no? -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CAH_OBidTGH=Dof=4v7eXRrdfy2oOLUA=5BexHupCS4_=qto...@mail.gmail.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
Hi Stan, Am Montag, 27. August 2012 schrieb Stan Hoeppner: On 8/27/2012 8:27 AM, Stefan Monnier wrote: I run an SSD on my MCP61P so lack of NCQ has no impact whatsoever--SSD's have no moving parts, and all seeks are instantaneous. While I haven't heard of NCQ improving read speed of SSDs, it can have a significant positive impact on write speed for SSDs. Some SSD controllers, such as the later Sanforce models, do benefit from NCQ with some server oriented workloads, little to none with others. Controllers such as the Indilinx, Jmicron, and Samsung don't benefit from NCQ at all, with any workload. For a desktop user workload, there will be no noticeable performance difference, because such applications don't do parallel IO. Are you sure about 1) desktop applications I/O behavior and 2) NCQ? 1) I see noticeable difference for my Intel SSD 320 with different iodepth values in fio job. And as to my current understanding more than one single-threaded I/O generating desktop application can easily run at some given point of time. And then Nepomuk desktop search accesses the SSD disk with up to 10 threads at times excluding the Virtuoso database server with its 5,4 GiB database – no kidding, thats for real. Then there is Akonadi with the PostgreSQL database, KMail, Iceweasel with cache and sqlite3 database and whatnot. As sync I/O calls are synchronous on the syscall side but the kernel is still free to schedule the requests in bigger batches on the lower levels of the I/O stuck, I´d expect that desktop workloads can cause similar effects as somewhat higher I/O depths as one. Indeed running fio with numjobs=64 instead of iodepth=64 gives similar results with that Intel SSD except for higher context switch rate and CPU usage. (I can dig out those results if wanted.) 2) I am not sure about NCQ tough. I´d never disabled it on my tests. The Intel SSD 320 reports a queue depth of 32 with hdparm -I. It may not make much of a difference, I don´t know. Your seek time argument makes sense to me. But then I thought that the SSD firmware may have bigger chances to combine requests into erase block sized units when it gets more data to deal with at once. So it may make some difference for writes. But I could be completely off track with my assumption and I bet only a test will show. SSD firmwares are like big black interesting and fascinating boxes to me ;) Thanks, -- Martin 'Helios' Steigerwald - http://www.Lichtvoll.de GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7 -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/201208282101.45190.mar...@lichtvoll.de
OT: SSD, NCQ and I/O depth (was: Re: [SOLVED] Is my processor 32-bit or 64-bit?)
Sorry that I post twice, but in hindsight I thought I´d give this a more descriptive subject. Please reply to this second post. Hi Stan, Am Montag, 27. August 2012 schrieb Stan Hoeppner: On 8/27/2012 8:27 AM, Stefan Monnier wrote: I run an SSD on my MCP61P so lack of NCQ has no impact whatsoever--SSD's have no moving parts, and all seeks are instantaneous. While I haven't heard of NCQ improving read speed of SSDs, it can have a significant positive impact on write speed for SSDs. Some SSD controllers, such as the later Sanforce models, do benefit from NCQ with some server oriented workloads, little to none with others. Controllers such as the Indilinx, Jmicron, and Samsung don't benefit from NCQ at all, with any workload. For a desktop user workload, there will be no noticeable performance difference, because such applications don't do parallel IO. Are you sure about 1) desktop applications I/O behavior and 2) NCQ? 1) I see noticeable difference for my Intel SSD 320 with different iodepth values in fio job. And as to my current understanding more than one single-threaded I/O generating desktop application can easily run at some given point of time. And then Nepomuk desktop search accesses the SSD disk with up to 10 threads at times excluding the Virtuoso database server with its 5,4 GiB database – no kidding, thats for real. Then there is Akonadi with the PostgreSQL database, KMail, Iceweasel with cache and sqlite3 database and whatnot. As sync I/O calls are synchronous on the syscall side but the kernel is still free to schedule the requests in bigger batches on the lower levels of the I/O stuck, I´d expect that desktop workloads can cause similar effects as somewhat higher I/O depths as one. Indeed running fio with numjobs=64 instead of iodepth=64 gives similar results with that Intel SSD except for higher context switch rate and CPU usage. (I can dig out those results if wanted.) 2) I am not sure about NCQ tough. I´d never disabled it on my tests. The Intel SSD 320 reports a queue depth of 32 with hdparm -I. It may not make much of a difference, I don´t know. Your seek time argument makes sense to me. But then I thought that the SSD firmware may have bigger chances to combine requests into erase block sized units when it gets more data to deal with at once. So it may make some difference for writes. But I could be completely off track with my assumption and I bet only a test will show. SSD firmwares are like big black interesting and fascinating boxes to me ;) Thanks, -- Martin 'Helios' Steigerwald - http://www.Lichtvoll.de GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7 -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/201208282103.17852.mar...@lichtvoll.de
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Sun, 26 Aug 2012 09:48:27 -0400 (EDT), Stephen Powell wrote: What I meant was that I may buy the new mobo, processor, and RAM that you suggested and put it in this second machine I'm talking about. Well, I've looked into it, and that's not going to work. The mobo you suggested has stuff on it for which there are no connector holes on the back of the case. The built-in connector holes on the back of the case support A/C power (of course), PS/2 mouse, PS/2 keyboard (small-size PS/2 connectors in both cases), 2 USB ports, 2 serial ports, and one parallel port. In addition, there are knock-out plugs for what looks like a joystick/MIDI port and what looks like line-in, line-out, and microphone connectors for on-board sound. I would have to knock out the metal plugs, but at least the case can handle a mobo with on-board sound. But there are no connector holes for VGA output, DVI output, or an Ethernet jack for a network connection, nor are there knock-out plugs for the same. The power supply connector to the mobo is a 20-pin connector. The case is just not designed to handle modern mobos. I can't even identify the mobo. With no RAM, the BIOS won't initialize. It just beeps at me (about 1 second on and 1 second off). So the BIOS won't tell me what the mobo is. I searched in vain for some form of identification on the mobo silk screen itself, but all I could find was Made in China. Therefore, I don't know what kind of memory it needs. I'm afraid this other system is going to be good for nothing except possibly to be cannibalized for parts. A clue to its age: it has a couple of ISA (or possibly EISA) bus slots on it, in addition to PCI bus slots. And the video card appears to be VESA local bus. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/253085795.713939.1346184701634.javamail.r...@md01.wow.synacor.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/28/2012 12:43 PM, shawn wilson wrote: On Tue, Aug 28, 2012 at 4:04 AM, Stan Hoeppner s...@hardwarefreak.com wrote: It's memory bandwidth of 20GB/s was many times higher than any x86 server at that time as they all used a single P6 bus, with only 1GB/s bandwidth. 20GB/s is peanuts today given just two channels of DDR3-1333 have just over 20GB/s, but back then, in the late 1990s, this was huge. This CMP design also allowed assigning different amounts of memory to each of the hosts, with the firmware and custom crossbar chipset setting up the fences in the physical memory map. Individual PCI buses and IO devices could be assigned to any of the partitioned servers. In the first models, a console module had to be installed which included VGA, KB, mouse ports and each was controlled via a KVM switch. Later models had a much more intelligent solution in the form of a single system controller. This also facilitated the ability to cluster multiple sets of two physical hosts (up to 4 clusters per server) within a single server using system memory as the cluster interconnect, with latency thousands of times lower and bandwidth thousands of times higher than the fastest network interconnects at that time, this became immensely popular with Unisys customers, many running multiple clustered MS SQL servers within one ES7000 mainframe. wasn't the 20GB/s infiniband introduced in the late 90s / early 2k? The fist Infiniband hardware hit the market in the early 2000s. But its data rate was only 2Gbps, or ~200MB/s, one way. This is Infiniband 1x. That's 100 times less bandwidth than the memory system in the 32-way ES7000. And the latency is over 100 times higher. Now take into account that with memory based networking, all that is required to send a packet is a write to a memory location, and to receive it, all that is required is to read that memory location. that should about measure up to what you're describing, but with tons more scalability, no? Absolutely not. 4x QDR Infiniband is the most popular type in wide use today. Its one way bandwidth is only 4GB/s, and its one way latency is still many tens of times greater than the in-memory networking in the ES7000 we're discussing. Infiniband 12x QDR is currently used for switch-switch backbone links and node links in some HPC clusters, and has a one way data rate of 96Gbps, or 12GB/s, only 60% of the ES7000 memory bandwidth, while its latency is still some 10 times greater. The absolute fastest type of Infiniband is currently 12x EDR, with a one way data rate of 300Gbps, or 37.5GB/s. And just as some of the Internet's fastest backbone links at 10 Terabits/sec have far more bandwidth that the memory subsystem of a high end parallel server of over a decade ago, so does today's fastest implementation of Infiniband. This shouldn't be surprising. Just as it shouldn't be surprising than any AMD socket AM3 or better system has more memory bandwidth than the decade+ old high end server we're discussing. Technology doesn't sit still. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/503d275d.1020...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
I run an SSD on my MCP61P so lack of NCQ has no impact whatsoever--SSD's have no moving parts, and all seeks are instantaneous. While I haven't heard of NCQ improving read speed of SSDs, it can have a significant positive impact on write speed for SSDs. Stefan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/jwvbohwlb48.fsf-monnier+gmane.linux.debian.u...@gnu.org
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/27/2012 8:27 AM, Stefan Monnier wrote: I run an SSD on my MCP61P so lack of NCQ has no impact whatsoever--SSD's have no moving parts, and all seeks are instantaneous. While I haven't heard of NCQ improving read speed of SSDs, it can have a significant positive impact on write speed for SSDs. Some SSD controllers, such as the later Sanforce models, do benefit from NCQ with some server oriented workloads, little to none with others. Controllers such as the Indilinx, Jmicron, and Samsung don't benefit from NCQ at all, with any workload. For a desktop user workload, there will be no noticeable performance difference, because such applications don't do parallel IO. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/503ba10f.4010...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Mon, 27 Aug 2012 01:42:54 -0400 (EDT), Stan Hoeppner wrote: Stephen Powell wrote: I know enough to be dangerous, but not always enough to be competent. That's why I opened this thread in the first place. In this case the solution is as simple as downloading and reading the manual for your board: http://ph.academicdirect.ro/ISB_SE7500CW2.pdf The memory specs are on page 17. Stan, you are a wealth of information. I didn't search for the motherboard manual because, for some reason, I didn't think they were published online. Maybe I did an extensive search in the past for a motherboard or chipset manual and found nothing; so I already had it in my head that such things weren't available. But in this case I am happy to be proved wrong. By the way, there's something I don't understand. A 32-bit processor can only access 4G of real (extended) memory, right? So why are there motherboards available for 32-bit processors that support installing more than 4G of RAM? What good is memory that the processor can't address? -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/619145302.692624.1346109143744.javamail.r...@md01.wow.synacor.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
Stephen Powell writes: So why are there motherboards available for 32-bit processors that support installing more than 4G of RAM? http://en.wikipedia.org/wiki/Physical_Address_Extension -- John Hasler -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87bohvkg6y@thumper.dhh.gt.org
Re: [SOLVED] Is my processor 32-bit or 64-bit?
Stephen Powell zlinux...@wowway.com writes: On Mon, 27 Aug 2012 01:42:54 -0400 (EDT), Stan Hoeppner wrote: Stephen Powell wrote: I know enough to be dangerous, but not always enough to be competent. That's why I opened this thread in the first place. In this case the solution is as simple as downloading and reading the manual for your board: http://ph.academicdirect.ro/ISB_SE7500CW2.pdf The memory specs are on page 17. Stan, you are a wealth of information. I didn't search for the motherboard manual because, for some reason, I didn't think they were published online. Maybe I did an extensive search in the past for a motherboard or chipset manual and found nothing; so I already had it in my head that such things weren't available. But in this case I am happy to be proved wrong. By the way, there's something I don't understand. A 32-bit processor can only access 4G of real (extended) memory, right? So why are there motherboards available for 32-bit processors that support installing more than 4G of RAM? What good is memory that the processor can't address? There's been a Physical Address Extension for 32-bit x86 systems for a while now, which supports more than 32 bits of physical address. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1b1uirakdd@pfeifferfamily.net
Re: [SOLVED] Is my processor 32-bit or 64-bit?
Stephen Powell wrote: By the way, there's something I don't understand. A 32-bit processor can only access 4G of real (extended) memory, right? So why are there motherboards available for 32-bit processors that support installing more than 4G of RAM? What good is memory that the processor can't address? With PAE (physical address extensions) the processor *can* address more than 4G of ram. A single process is still limited to 32-bits which usually works out effectively to 3G of ram but the operating system can make use of more than this. It can be used for filesystem buffer cache and for multiple 3G programs. A machine with 6G of ram for example could run two 3G program at the same time and hold them both in memory without swapping. Or run one 3G program and still have 3G for the system to use in filesystem buffer cache. With PAE having more than 4G of memory is quite useful. Using PAE does have a small performance impact. It slows things down by 2%-3% in my use cases. But the increase in ram for buffers usually more than makes up for the differences. Bob signature.asc Description: Digital signature
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Tue, Aug 28, 2012 at 1:34 AM, Bob Proulx b...@proulx.com wrote: Stephen Powell wrote: By the way, there's something I don't understand. A 32-bit processor can only access 4G of real (extended) memory, right? So why are there motherboards available for 32-bit processors that support installing more than 4G of RAM? What good is memory that the processor can't address? With PAE (physical address extensions) the processor *can* address more than 4G of ram. A single process is still limited to 32-bits which usually works out effectively to 3G of ram but the operating system can make use of more than this. It can be used for filesystem buffer cache and for multiple 3G programs. A machine with 6G of ram for example could run two 3G program at the same time and hold them both in memory without swapping. Or run one 3G program and still have 3G for the system to use in filesystem buffer cache. With PAE having more than 4G of memory is quite useful. Using PAE does have a small performance impact. It slows things down by 2%-3% in my use cases. But the increase in ram for buffers usually more than makes up for the differences. iirc, pae is only 48 bits too. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CAH_OBifAaW=m2tqojv6ksybnbp2bqecm1cf4fy8c9y9aqzk...@mail.gmail.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Sat, 25 Aug 2012 23:44:50 -0400 (EDT), Stan Hoeppner wrote: Search for GeForce 6150. That's the integrated GPU. Wikipedia tells you this. My mistake. I thought MCP61 was the chipset name. The Wikipedia article I found also indicated that MCP61 ... introduced a bug in the SATA NCQ implementation. As a result, Nvidia employees have contributed code to disable NCQ operations under Linux. But I have no idea what that means, or whether or not it would be relevant in my case. If you need parts for old systems, you simply can't be Ebay. Just about anything you could need is available, and usually really cheap. The older the tech the cheaper. The opposite is normally true when buying from computer retailers. In this case the older it is the more they want for it, if they even have it or can get it. They treat such parts as collectable antique furniture, price wise. What I meant was that I may buy the new mobo, processor, and RAM that you suggested and put it in this second machine I'm talking about. But yes, that's true about E-bay (and other sites too). But the key is you simply *must* know *exactly* what you want/need. Retail vendors, such as oempcworld.com, usually have a web site where you can identify your system somehow and then they will tell *you* what you need (or at least what they recommend). On E-bay, it's, I have this for sale. Do you want it or not? And often, for me, the answer is, I'm not sure. If I buy the wrong thing, and that has happened before, I'm usually out the money and stuck with something I can't use. The terms are often all sales final. I simply am not a hardware expert. I know enough to be dangerous, but not always enough to be competent. That's why I opened this thread in the first place. I went ahead and bought that lot of 14 DIMMs from E-bay; but without your assurances that it would work in my system, I would not have had enough confidence to buy it. Thanks for your help. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1330577568.661637.1345988907614.javamail.r...@md01.wow.synacor.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/26/2012 8:48 AM, Stephen Powell wrote: On Sat, 25 Aug 2012 23:44:50 -0400 (EDT), Stan Hoeppner wrote: Search for GeForce 6150. That's the integrated GPU. Wikipedia tells you this. My mistake. I thought MCP61 was the chipset name. It can be a bit confusing I guess. Here's a pre-release article that explains the MCP61 family in some detail: http://www.dailytech.com/article.aspx?newsid=3742 It's not a high end chipset by any means, but more than plenty for any regular desktop duty. One of the systems I rebuilt with the Foxconn MCP61P is running Win7 and is plenty quick with Aero enabled. The Wikipedia article I found also indicated that MCP61 ... introduced a bug in the SATA NCQ implementation. As a result, Nvidia employees have contributed code to disable NCQ operations under Linux. But I have no idea what that means, or whether or not it would be relevant in my case. With some systems and some workloads, NCQ can give a little performance boost. NCQ allows a rusty drive to change the order of sector accesses in order to optimize head seek latency. With normal desktop application usage you'd never notice a difference w/NCQ vs no NCQ. I run an SSD on my MCP61P so lack of NCQ has no impact whatsoever--SSD's have no moving parts, and all seeks are instantaneous. If you need parts for old systems, you simply can't be Ebay. Just about anything you could need is available, and usually really cheap. The older the tech the cheaper. The opposite is normally true when buying from computer retailers. In this case the older it is the more they want for it, if they even have it or can get it. They treat such parts as collectable antique furniture, price wise. What I meant was that I may buy the new mobo, processor, and RAM that you suggested and put it in this second machine I'm talking about. But yes, that's true about E-bay (and other sites too). But the key is you simply *must* know *exactly* what you want/need. Well, sure. Retail vendors, such as oempcworld.com, usually have a web site where you can identify your system somehow and then they will tell *you* what you need (or at least what they recommend). On E-bay, it's, I have this for sale. Do you want it or not? And often, for me, the answer is, I'm not sure. If I buy the wrong thing, and that has happened before, I'm usually out the money and stuck with something I can't use. The terms are often all sales final. I simply am not a hardware expert. I know enough to be dangerous, but not always enough to be competent. That's why I opened this thread in the first place. In this case the solution is as simple as downloading and reading the manual for your board: http://ph.academicdirect.ro/ISB_SE7500CW2.pdf The memory specs are on page 17. I went ahead and bought that lot of 14 DIMMs from E-bay; but without your assurances that it would work in my system, I would not have had enough confidence to buy it. Thanks for your help. You shouldn't need my assurance. The advertised specs match those in the manual. The seller has a 99.4% rating. A browse of his Ebay store indicates that he only sells computer hardware. Unless he accidentally ships you the wrong parts then you should be fine. Also note that at the bottom of the auction it states these DIMMs are covered by the seller's 7 day no DOA warranty. This text takes precedence over the Returns policy in the sale details box top of page which states no returns. Thus, I'd recommend you test all of the modules as soon as you get them. You can make 2 sets of 4 of the low profile modules, then test 4 of the high profile units. Test the last two units observing the slot population guidelines in the manual so you're in dual channel mode. Use memtest86--don't just assume they work because the box posts. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/503b08de.5000...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Fri, 24 Aug 2012 05:45:23 -0400 (EDT), Stan Hoeppner wrote: On 8/23/2012 7:28 PM, Stephen Powell wrote: Hmm. I'm not even sure if this is the right kind of memory. It is. OK, if you say so. I guess oempcworld.com is trying to sell me faster memory than I really need. I was confused by PC1600 vs PC2100. Under the photo: Actual item may differ. That photo shows a standard 184 pin non ECC desktop DIMM. Right. I understand they used a generic picture. The nVidia MCP61 SB+GPU single chip chipset on this board hit the market in late 2006 and has been out of production for at least 2 years, boards using it out of production for a year. Units still in the channel are simply those that haven't sold. This is the reason this is the cheapest socket AM3 board Newegg offers. Sometimes you forget you and I think alot alike. I'd never recommend anything bleeding edge to you Stephen. I could find no evidence of support for this chip in the Debian package descriptions (on-line web pages) for nvidia-kernel-source, nvidia-kernel-legacy-173xx-source, or nvidia-kernel-legacy-96xx-source for the Squeeze release. I could not find MCP61 anywhere. Of course, the nvidia-kernel-legacy-71xx-source package exists only in Sid because of build problems. It doesn't work with any release past Lenny. But I searched it too, just for grins, and couldn't find MCP61 on that web page either. By the way, that brings up a pet peeve of mine. I should probably open another thread for it. The Wheezy versions of the above Debian packages (i.e. the on-line web pages which describe them) have *no* description of which chipsets they support. How is one supposed to know which package one needs if one is running Wheezy? The 2D chip is too slow. If you want to use a GUI desktop and a modern browser, visit modern complex websites, play flash videos etc, this system won't be very snappy. That depends on one's level of expectations. I'm currently using a 1 GHz single core, single thread, 32-bit processor and only 512M of RAM. The video driver is nouveau, and for this chipset, it apparently does not support 3D acceleration. So, believe it or not, I expect that the built- in Intel video will be an improvement over what I've got now. If not, installing a separate video card is always an option. You'll also likely need to add a sound card. Actually, no. There is no built-in sound chip on the mobo, but there is a sound card installed in an expansion slot. Apparently, the previous owner was using it as a desktop system, even though it was originally intended as a server. Now, if you plan to use it as a headless server, spend the $50 on the memory and use it as is. Should be fine for some light duty stuff. Actually, I just might do both. I have another computer that someone gave to me that has no RAM at all on it. Apparently the previous owner removed the RAM from the old computer in order to use it in his new computer before he gave me the old computer. Maybe I'll just replace the mobo in it. I need to do more research first though. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/671039830.648449.1345905257815.javamail.r...@md01.wow.synacor.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Fri, 24 Aug 2012 09:37:17 -0400 (EDT), Stan Hoeppner wrote: On 8/24/2012 6:54 AM, Stephen Powell wrote: OK, now we're getting somewhere. There is an eight-pin connector coming from the power supply that plugs directly into the motherboard, in addition to a 24-pin connector coming from the power supply which also plugs directly into the motherboard. Yes, I covered this is some detail in a section you snipped. ;) Actually, I think the information to which you refer was not something that I snipped, but rather something which was in another post, one that I accidentally deleted and had not read at the time I wrote the above. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/934720309.648628.1345905631444.javamail.r...@md01.wow.synacor.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/24/2012 1:53 PM, Bob Proulx wrote: In actuality I have only used these adapters on low power Atom motherboards. I am down in the 20 watts of power envelope area. So for me the single rail is more than enough. 4 pin aux CPU power plug on at Atom board? That's just silly. My now retired Barton system had a TDP of 76.8W and used a standard 20 pin ATX connector, drawing all power across board traces. Lasted 8 years. This Atom board you mention makes me think it's simply easier to design mobos that use aux power instead of pulling from the main plug, so they're doing it even when it's not necessary. I just saw a rather expensive 700W supply in a friend's system burn up shooting sparks out the back fan and burning a large hole in the circuit board. Back in '95 I had to steal the fan out of my bench supply to put in a customer's machine. It was an old PSU so we didn't bother replacing the fan, strictly using it for quick/dirty testing and what not from that point, only running for short durations. You know where this is going... I had a new build parts set on the bench and had fired it up for a quick smoke test. I got called away for just a sec and left it running. After the bosses wife had been chatting me up for about 5 minutes, there's a loud pop accompanied by a simultaneous brilliant light momentarily blinding everyone in the room. The PSU is showering my bench with sparks much like a thermite reaction. The breaker didn't trip for some reason. I dashed to the bench, and whilst avoiding the shower of sparks, ripped the power cord from the wall socket. It continued shooting sparks for another ~30 seconds, apparently due to the intense heat buildup that led to the ignition. Everything was so badly burned and welded together it was impossible to determine which component overheated first and let go. I'd suppose it had to be one of the MOSFETs in the switching circuit. If anyone has an old PSU and wants some cheap kicks, invite some friends over, remove the fan(s), take it outside and get 15 feet away. Plug it into an extension cord, power it up, sit back in the lawn chairs with a cold one and wait patiently for 5-20 minutes (some pop sooner others later). If you want it to pop faster remove the heatsinks from the MOSFETs. The spark shower won't be as intense in absence of chunks of the aluminum heatsinks being liquified, but it'll ignite more quickly. If you want maximum spark show, completely remove the board from the chassis and sit it on plastic. You'll probably want to be more than 15 feet away if you go this route. DISCLAIMER: I am not responsible nor liable for any personal injury or property damage resulting from following these suggestions. The individual engaging in such activity accepts full responsibility for any/all consequences. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/50399339.7070...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/25/2012 9:34 AM, Stephen Powell wrote: OK, if you say so. I guess oempcworld.com is trying to sell me faster memory than I really need. I was confused by PC1600 vs PC2100. IIRC PC1600 is DDR200. PC2100 is DDR266. Identical memory certified for different frequencies. That Intel SE7500 will work with either frequency. I could find no evidence of support for this chip in the Debian package descriptions (on-line web pages) for nvidia-kernel-source, nvidia-kernel-legacy-173xx-source, or nvidia-kernel-legacy-96xx-source for the Squeeze release. I could not find MCP61 anywhere. Of course, the nvidia-kernel-legacy-71xx-source package exists only in Sid because of build problems. It doesn't work with any release past Lenny. But I searched it too, just for grins, and couldn't find MCP61 on that web page either. Search for GeForce 6150. That's the integrated GPU. Wikipedia tells you this. That depends on one's level of expectations. I'm currently using a 1 GHz single core, single thread, 32-bit processor and only 512M of RAM. The video driver is nouveau, and for this chipset, it apparently does not support 3D acceleration. So, believe it or not, I expect that the built- in Intel video will be an improvement over what I've got now. If not, installing a separate video card is always an option. The SE7500 does not an Intel GPU. It doesn't have a GPU. It's a 2D ATI chip with limited 3D support. In fact this was the first chip series to which ATI added any 3D features. The ATI Rage series of chips were released between 1995 and 1999. That should tell you something about the performance, and the 75MHz clock, ouch. From the Intel board manual: The Intel® Server Board SE7500CW2 provides an ATI Rage XL PCI graphics accelerator, along with 8 MB of video SDRAM and support circuitry for an embedded SVGA video subsystem. From: http://en.wikipedia.org/wiki/ATI_Rage 75 MHz core 4, 8, and 16 MB 100 MHz SGRAM/WRAM memory 800 MB/s memory bandwidth DirectX 6.0 Rage XL was a low-cost RAGE Pro-based solution. As a low-power solution with capable 2D-acceleration, the chip was used on many low-end graphics cards. It was also seen on Intel motherboards, as recently as 2004, and was still used in 2006 for server motherboards. The Rage XL has been succeeded by the ATI ES1000 for server use. The chip was basically a die-shrunk Rage Pro, optimized to be very inexpensive for solutions where only basic graphics output was necessary. Actually, no. There is no built-in sound chip on the mobo, but there is a sound card installed in an expansion slot. Apparently, the previous owner was using it as a desktop system, even though it was originally intended as a server. Cool. Actually, I just might do both. I have another computer that someone gave to me that has no RAM at all on it. Apparently the previous owner removed the RAM from the old computer in order to use it in his new computer before he gave me the old computer. Maybe I'll just replace the mobo in it. I need to do more research first though. If you need parts for old systems, you simply can't be Ebay. Just about anything you could need is available, and usually really cheap. The older the tech the cheaper. The opposite is normally true when buying from computer retailers. In this case the older it is the more they want for it, if they even have it or can get it. They treat such parts as collectable antique furniture, price wise. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/50399bb2.6070...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
Stan Hoeppner wrote: On 8/24/2012 1:53 PM, Bob Proulx wrote: In actuality I have only used these adapters on low power Atom motherboards. I am down in the 20 watts of power envelope area. So for me the single rail is more than enough. 4 pin aux CPU power plug on at Atom board? That's just silly. Agreed. But so it is. I found a listing for the product. I got it about $15 cheaper than this listing however. Add 4G of ram for $22 and it isn't a bad system. It isn't a performance machine. So don't expect it to be a video transcoder or gaming machine. But for the power envelope it is great. http://www.newegg.com/Product/Product.aspx?Item=N82E16813121442 My now retired Barton system had a TDP of 76.8W and used a standard 20 pin ATX connector, drawing all power across board traces. Lasted 8 years. This Atom board you mention makes me think it's simply easier to design mobos that use aux power instead of pulling from the main plug, so they're doing it even when it's not necessary. Probably. Back in '95 I had to steal the fan out of my bench supply to put in a customer's machine. It was an old PSU so we didn't bother replacing the fan, strictly using it for quick/dirty testing and what not from that point, only running for short durations. You know where this is going... Good funny story! Thanks for sharing. If anyone has an old PSU and wants some cheap kicks, invite some friends I think that must be accompanied by CO2 soda pop bottle bombs for the complete experience. And possibly a potato gun to trigger the hang-fires. :-) Bob signature.asc Description: Digital signature
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/23/2012 7:28 PM, Stephen Powell wrote: On Thu, 23 Aug 2012 03:07:49 -0400 (EDT), Stan Hoeppner wrote: http://www.ebay.com/itm/Lot-of-14-PC1600-DDR-200-Registered-ECC-1GB-Server-Memory-Micron-Samsung-/130718246446?pt=US_Memory_RAM_hash=item1e6f6a262e Hmm. I'm not even sure if this is the right kind of memory. It is. I only need 4. What would I do with the other 10? If 14 used sticks is $50 and 4 new sticks is $120, does it matter? Keep em as spares, make key chains, whatever. I went to the web site of an outfit I've dealt with before, http://www.oempcworld.com, and I found a way to specify my system by entering the motherboard number. I told it I had an Intel SE7500CW2. Here's the page it took me to: http://www.oempcworld.com/configurator/configurator.php?mch=SE7500CW2mfr=Intelmdn=SE7500+Mainboard+Series Under the photo: Actual item may differ. That photo shows a standard 184 pin non ECC desktop DIMM. They are charging about $30 per DIMM, so four of them would be about $120. The lot of 14 from Ebay is the only way to go here. $50 for old RAM or $110 few all new guts? $110 gets you a new Foxconn AM3 mobo, 2.8GHz 1MB L2 64bit 45 watt single core AMD retail CPU, and 4GB DDR3-1333 dual channel RAM-- 6.6x the memory bandwidth of the Netburst Xeon. http://www.newegg.com/Product/Product.aspx?Item=N82E16819103888 http://www.newegg.com/Product/Product.aspx?Item=N82E16813186189 http://www.newegg.com/Product/Product.aspx?Item=N82E16820148541 I've used this combo to refurb 2 old machines now, including the machine from which I've typing this. No problems so far with exactly one year on this one. I've got a dual core Regor 3GHz 2x1MB L2 in this box. If a single core 2.8 64bit Sempron is insufficient for your workload, add $25 for a 65 watt 3.2GHz dual core AthlonII X2, $135 total: http://www.newegg.com/Product/Product.aspx?Item=N82E16819103873 Hmm. That's worth mulling over. But I generally like to get hardware when it's at least 3 or 4 years old. That way, I can usually run Debian stable and have all the device drivers work with no problems. I worry particularly about the mobo with an on-board bleeding edge Nvidia video chipset. The nVidia MCP61 SB+GPU single chip chipset on this board hit the market in late 2006 and has been out of production for at least 2 years, boards using it out of production for a year. Units still in the channel are simply those that haven't sold. This is the reason this is the cheapest socket AM3 board Newegg offers. Sometimes you forget you and I think alot alike. I'd never recommend anything bleeding edge to you Stephen. This thing is about 10 years old, but since it was originally intended to be used as a high-end server, it should make a decent desktop system even today. The 2D chip is too slow. If you want to use a GUI desktop and a modern browser, visit modern complex websites, play flash videos etc, this system won't be very snappy. You'll also likely need to add a sound card. This is decidedly an entry level server. Entry is 2-4 sockets. Midrange is 8-16 sockets. High end is 16+ sockets. There have only been a handful of high end x86 servers produced throughout history: the 32-way Data General Aviion PPro, the 32-way Unisys ES7000 P3 Xeon, both long discontinued, and the currently shipping 256-way SGI Altix UV 4/8/10 core Xeon, the very highest of the high end. It's the only shared memory system of any ISA with over 128 sockets. As long as the PSU has the 4-pin CPU power plug, and it should being a Xeon board, you shouldn't need to replace anything else. And you've basically got a brand new system, sans drives, for $110-135. I don't see the 4-pin CPU power plug to which you refer coming out of the power supply. (Yes, I finally broke down and took the cover off.) Of course, it does have several spare 4-pin power connectors designed for peripherals, such as hard disks, CD-ROM drives, etc. But I suspect you are referring to something smaller. Found a good pic of this board. It has an 8 pin aux +12v power connector, not 4 pin. It may be unused on your system. I.e. the person who built the box didn't use a PSU with the 8 pin aux output as only one CPU was installed and most of the PCI slots are empty. I didn't read the manual thoroughly. You could probably get away without connecting the 4 pin aux CPU power on the Foxconn board if using a 65w or lower CPU. I've never tried it. But I don't find anything in the manual that says the board won't post with it disconnected. Many newer boards won't power up without it connected and their docs say so in bold print. This is a safety feature to keep folks with 80+ watt CPUs from burning up the board traces and/or smoking the 24pin +12V wires due to excessive current draw. Worse case scenario if it must be connected, you also buy a $15 PSU with the 4 pin aux CPU +12V output. It's still a big win over spending $120 to add memory to an old slow
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/24/2012 12:57 AM, Bob Proulx wrote: Stephen Powell wrote: Stan Hoeppner wrote: As long as the PSU has the 4-pin CPU power plug, and it should being a Xeon board, you shouldn't need to replace anything else. And you've basically got a brand new system, sans drives, for $110-135. I don't see the 4-pin CPU power plug to which you refer coming out of the power supply. (Yes, I finally broke down and took the cover off.) Of course, it does have several spare 4-pin power connectors designed for peripherals, such as hard disks, CD-ROM drives, etc. But I suspect you are referring to something smaller. Here is a reference for you: http://en.wikipedia.org/wiki/Power_supply_unit_(computer)#ATX12V_standard Most newer motherboards now require this addtional power connector. But if your power supply does not provide one then you can add an adaptor and convert one of the 4-pin power connectors to the ATX12V 4-pin motherboard power connector. That works fine. I have done that on a couple of systems. This method is usually fine for lower wattage CPUs. Be aware that using this method provides only one additional +12V lead (wire) and one ground from the PSU. The 4 pin aux CPU power standard provides two +12V leads and two grounds. The new 8 pin aux standard provides 4 +12V leads and 4 grounds. If running an 80+ watt CPU I recommend against this converter strategy as you'll have only 1/2 of the aux +12V conductors needed to safely carry the current load. You may sneak by at 80W, but at 95W you're tickling the dragon, and with a 135W CPU you will fry some wires, ruining your PSU, and likely the mobo as well. Decent PSUs with the proper aux connectors are really inexpensive. Why gamble? -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/5037546c.5030...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Fri, 24 Aug 2012 06:16:12 -0400 (EDT), Stan Hoeppner wrote: Stephen Powell wrote: I don't see the 4-pin CPU power plug to which you refer coming out of the power supply. (Yes, I finally broke down and took the cover off.) Of course, it does have several spare 4-pin power connectors designed for peripherals, such as hard disks, CD-ROM drives, etc. But I suspect you are referring to something smaller. ... The new 8 pin aux standard provides 4 +12V leads and 4 grounds. ... OK, now we're getting somewhere. There is an eight-pin connector coming from the power supply that plugs directly into the motherboard, in addition to a 24-pin connector coming from the power supply which also plugs directly into the motherboard. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1911564965.627983.1345809241358.javamail.r...@md01.wow.synacor.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/24/2012 6:54 AM, Stephen Powell wrote: On Fri, 24 Aug 2012 06:16:12 -0400 (EDT), Stan Hoeppner wrote: Stephen Powell wrote: I don't see the 4-pin CPU power plug to which you refer coming out of the power supply. (Yes, I finally broke down and took the cover off.) Of course, it does have several spare 4-pin power connectors designed for peripherals, such as hard disks, CD-ROM drives, etc. But I suspect you are referring to something smaller. ... The new 8 pin aux standard provides 4 +12V leads and 4 grounds. ... OK, now we're getting somewhere. There is an eight-pin connector coming from the power supply that plugs directly into the motherboard, in addition to a 24-pin connector coming from the power supply which also plugs directly into the motherboard. Yes, I covered this is some detail in a section you snipped. ;) Your options should you choose to replace the guts... Newegg has an Athena Power adapter for $5 that will mate your PSU's 8 pin aux power plug to the 4 pin of the Foxconn board I recommended. Or, for $15 you can get a much better Foxconn desktop oriented mobo that includes the 8 pin aux CPU power connector, as well as Radeon HD4250 GPU, VGA, DVI and HDMI video outputs, 5 SATAII ports, an eSATA port, 8 channel audio, 1 PCIe x16 and two PCIe x1 slots, etc. Works with the same CPU and RAM I previously mentioned: http://www.newegg.com/Product/Product.aspx?Item=N82E16813186206 Given the 4-8x performance increase of new hardware, for little more than the cost of adding memory to the box, it's difficult to justify hanging onto the original innards. Of course, this has been the case for many years. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/5037838d.1000...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
Stan Hoeppner wrote: Bob Proulx wrote: Most newer motherboards now require this addtional power connector. But if your power supply does not provide one then you can add an adaptor and convert one of the 4-pin power connectors to the ATX12V 4-pin motherboard power connector. That works fine. I have done that on a couple of systems. This method is usually fine for lower wattage CPUs. Be aware that using this method provides only one additional +12V lead (wire) and one ground from the PSU. The 4 pin aux CPU power standard provides two +12V leads and two grounds. The new 8 pin aux standard provides 4 +12V leads and 4 grounds. I have two of those adapters. One uses one 4-pin power plug. The other uses two 4-pin power plugs. I take it that you would approve of the one that uses two but not the one that uses only one. :-) And now I could be cantankerous and plug those two power plugs into a single power feed wire that had two connectors? Obviously for more power I would at the least try to use a second power supply rail and connect one each to two different rails. So the one with two 4-pin plugs could be plugged into two different supply rails and everything would meet spec just fine. Or I could plug it in oppositely and use a cpu that needs excessive 12V current and over stress things. I am sorry but I am just being cantankerous in this paragraph. If running an 80+ watt CPU I recommend against this converter strategy as you'll have only 1/2 of the aux +12V conductors needed to safely carry the current load. You may sneak by at 80W, but at 95W you're tickling the dragon, and with a 135W CPU you will fry some wires, ruining your PSU, and likely the mobo as well. Decent PSUs with the proper aux connectors are really inexpensive. Why gamble? If you are running that much power then I think upgrading the power supply is the best answer. In actuality I have only used these adapters on low power Atom motherboards. I am down in the 20 watts of power envelope area. So for me the single rail is more than enough. On every other system I have simply upgraded to a newer power supply. The new supply has the newer power leads and no adapters are needed. Being new I feel that it should reset the age clock on them. Being new they tend to be more efficient and draw less total power. They have become extremely cheap (and I say cheap instead of inexpensive) and replacing the power supply for $15 and keeping a couple of spares around to replace them as they need it has become very cost effective. But even expensive supplies fail. I just saw a rather expensive 700W supply in a friend's system burn up shooting sparks out the back fan and burning a large hole in the circuit board. It appears that heat and vibration caused it to rub a part through a soft insulator and short a component against a frame member. That is by far the most violent failure I have seen for a long time! Most of the cheap ones fail by fan bearing failure and without a fan it simply gets too hot and goes into thermal shutdown. Be sure to use oven hot pads when handling such items as they are extremely hot. Bob signature.asc Description: Digital signature
Re: [SOLVED] Is my processor 32-bit or 64-bit?
Stan Hoeppner wrote: You could probably get away without connecting the 4 pin aux CPU power on the Foxconn board if using a 65w or lower CPU. I've never tried it. But I don't find anything in the manual that says the board won't post with it disconnected. Many newer boards won't power up without it connected and their docs say so in bold print. This is a safety feature to keep folks with 80+ watt CPUs from burning up the board traces and/or smoking the 24pin +12V wires due to excessive current draw. On the low power Atom D525 it won't POST with the 4-pin cpu power disconnected. I tried it. Worse case scenario if it must be connected, you also buy a $15 PSU with the 4 pin aux CPU +12V output. It's still a big win over spending $120 to add memory to an old slow box. Now, if you plan to use it as a headless server, spend the $50 on the memory and use it as is. Should be fine for some light duty stuff. Agreed. I reuse a lot of older hardware too. Here is another general observation. I routinely measure the total power of a system and label it for reference later. I have been moving out some of the older power hungry machines and replacing them with newer greener lower power systems. Like the Atom which can make a nice low watt system very easily and has reasonable performance for many tasks. But that doesn't mean that all old hardware is high power. I am still using a Pentium 166MHz machine in a dedicated role because at 33 watts it is still doing the job I need it to do and it is lower power than most newer machines. I can beat 33 watts today (but at a significant cost) and will probably change it out soon but for example a new Foxconn machine I built up recently was 65 watts. Simply replacing old with new isn't always lower power. It must be measured to be sure. A previous generation core 2 duo of mine runs 150 watts. Machines in the high wattage envelope are pretty common. I still use that power hungry machine but I turn it off between uses. I also have an IBM workstation that burns 250 watts. That one is off most of the time. My main desktop for program development and email I have converted to a 40 watt low power Core 2 Quad system. For systems that I have running 24x7 I am now optimizing for a combination of total power and reasonable performance. Not peak gaming machine performance. Peak performance can pull a lot of power. The GPU is the newest single power hog in high performance systems. So for me I now have dedicated machines. Some run 24x7 and I try to make them as low of power as possible. Others that need peak performance and are going to draw high power I turn on for when I want to use them, use them, and then I turn them off when I am not using them. The idea of having one desktop that does it all is no longer viable for me. I try to optimize each machine for its task. Bob signature.asc Description: Digital signature
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/22/2012 10:16 PM, Henrique de Moraes Holschuh wrote: Now, my next step is to figure out what memory SIMMs to order. I'd like to install four 1G SIMMs, if they exist for this motherboard. But the devil is in the details. DIMMs not SIMMs. ;) SIMMs haven't been used in new systems for about 15 years. You probably won't find that registered ECC DDR200 through retail channels because of its age. If you do it'll likely be $100+/stick. Here's the best deal I found on Ebay: http://www.ebay.com/itm/Lot-of-14-PC1600-DDR-200-Registered-ECC-1GB-Server-Memory-Micron-Samsung-/130718246446?pt=US_Memory_RAM_hash=item1e6f6a262e $50 for old RAM or $110 few all new guts? $110 gets you a new Foxconn AM3 mobo, 2.8GHz 1MB L2 64bit 45 watt single core AMD retail CPU, and 4GB DDR3-1333 dual channel RAM-- 6.6x the memory bandwidth of the Netburst Xeon. http://www.newegg.com/Product/Product.aspx?Item=N82E16819103888 http://www.newegg.com/Product/Product.aspx?Item=N82E16813186189 http://www.newegg.com/Product/Product.aspx?Item=N82E16820148541 I've used this combo to refurb 2 old machines now, including the machine from which I've typing this. No problems so far with exactly one year on this one. I've got a dual core Regor 3GHz 2x1MB L2 in this box. If a single core 2.8 64bit Sempron is insufficient for your workload, add $25 for a 65 watt 3.2GHz dual core AthlonII X2, $135 total: http://www.newegg.com/Product/Product.aspx?Item=N82E16819103873 As long as the PSU has the 4-pin CPU power plug, and it should being a Xeon board, you shouldn't need to replace anything else. And you've basically got a brand new system, sans drives, for $110-135. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/5035d6c5.1090...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Thu, 23 Aug 2012 03:07:49 -0400 (EDT), Stan Hoeppner wrote: Stephen Powell wrote: Now, my next step is to figure out what memory SIMMs to order. I'd like to install four 1G SIMMs, if they exist for this motherboard. But the devil is in the details. DIMMs not SIMMs. ;) SIMMs haven't been used in new systems for about 15 years. Right. They are Dual In-line Memory Modules, not Single In-line Memory Modules. Old habits die hard. You probably won't find that registered ECC DDR200 through retail channels because of its age. If you do it'll likely be $100+/stick. Here's the best deal I found on Ebay: http://www.ebay.com/itm/Lot-of-14-PC1600-DDR-200-Registered-ECC-1GB-Server-Memory-Micron-Samsung-/130718246446?pt=US_Memory_RAM_hash=item1e6f6a262e Hmm. I'm not even sure if this is the right kind of memory. Plus, I only need 4. What would I do with the other 10? I went to the web site of an outfit I've dealt with before, http://www.oempcworld.com, and I found a way to specify my system by entering the motherboard number. I told it I had an Intel SE7500CW2. Here's the page it took me to: http://www.oempcworld.com/configurator/configurator.php?mch=SE7500CW2mfr=Intelmdn=SE7500+Mainboard+Series They are charging about $30 per DIMM, so four of them would be about $120. $50 for old RAM or $110 few all new guts? $110 gets you a new Foxconn AM3 mobo, 2.8GHz 1MB L2 64bit 45 watt single core AMD retail CPU, and 4GB DDR3-1333 dual channel RAM-- 6.6x the memory bandwidth of the Netburst Xeon. http://www.newegg.com/Product/Product.aspx?Item=N82E16819103888 http://www.newegg.com/Product/Product.aspx?Item=N82E16813186189 http://www.newegg.com/Product/Product.aspx?Item=N82E16820148541 I've used this combo to refurb 2 old machines now, including the machine from which I've typing this. No problems so far with exactly one year on this one. I've got a dual core Regor 3GHz 2x1MB L2 in this box. If a single core 2.8 64bit Sempron is insufficient for your workload, add $25 for a 65 watt 3.2GHz dual core AthlonII X2, $135 total: http://www.newegg.com/Product/Product.aspx?Item=N82E16819103873 Hmm. That's worth mulling over. But I generally like to get hardware when it's at least 3 or 4 years old. That way, I can usually run Debian stable and have all the device drivers work with no problems. I worry particularly about the mobo with an on-board bleeding edge Nvidia video chipset. This thing is about 10 years old, but since it was originally intended to be used as a high-end server, it should make a decent desktop system even today. As long as the PSU has the 4-pin CPU power plug, and it should being a Xeon board, you shouldn't need to replace anything else. And you've basically got a brand new system, sans drives, for $110-135. I don't see the 4-pin CPU power plug to which you refer coming out of the power supply. (Yes, I finally broke down and took the cover off.) Of course, it does have several spare 4-pin power connectors designed for peripherals, such as hard disks, CD-ROM drives, etc. But I suspect you are referring to something smaller. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/2073239832.67.1345768094127.javamail.r...@md01.wow.synacor.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
Stephen Powell wrote: Stan Hoeppner wrote: As long as the PSU has the 4-pin CPU power plug, and it should being a Xeon board, you shouldn't need to replace anything else. And you've basically got a brand new system, sans drives, for $110-135. I don't see the 4-pin CPU power plug to which you refer coming out of the power supply. (Yes, I finally broke down and took the cover off.) Of course, it does have several spare 4-pin power connectors designed for peripherals, such as hard disks, CD-ROM drives, etc. But I suspect you are referring to something smaller. Here is a reference for you: http://en.wikipedia.org/wiki/Power_supply_unit_(computer)#ATX12V_standard Most newer motherboards now require this addtional power connector. But if your power supply does not provide one then you can add an adaptor and convert one of the 4-pin power connectors to the ATX12V 4-pin motherboard power connector. That works fine. I have done that on a couple of systems. Bob signature.asc Description: Digital signature
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On 8/21/2012 9:04 PM, Stephen Powell wrote: On Tue, 21 Aug 2012 04:18:52 -0400 (EDT), Stan Hoeppner wrote: Yep. CPUID 0F27 makes this CPU a Prestonia Xeon, 130nm, in essence a Northwood P4, the only difference being the model#, CPUID, and branding. Intel introduced EM64T (x86-64) with the 90nm chips. This CPU is 32bit x86 only. Stan, Stan, the hardware man! I knew you'd know! Where did you find the information that correlates CPUIDs with processor characteristics? Simplicity. Google CPUID 0F27. First hit is: http://www.cpu-world.com/sspec/SL/SL6YH.html That tells us 0F27 is the Northwood core. I don't suppose you'd have a URL handy, now would you? Next time, I can check myself and not have to bother the list. Once we have the above info, Wikipedia tells us the Northwood chips were all 130nm, and that EM64T wasn't introduced until the 90nm series. The 2.4GHz/512KB/400MHz Prestonia CPUID 0F27 was released in April 2002. Nacona (90nm) with EM64T was introduced over two years later, in June 2004. -- Stan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/50349b36.6050...@hardwarefreak.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Wed, 22 Aug 2012 04:41:26 -0400 (EDT), Stan Hoeppner wrote: On 8/21/2012 9:04 PM, Stephen Powell wrote: Stan, Stan, the hardware man! I knew you'd know! Where did you find the information that correlates CPUIDs with processor characteristics? Simplicity. Google CPUID 0F27. First hit is: http://www.cpu-world.com/sspec/SL/SL6YH.html That tells us 0F27 is the Northwood core. I don't suppose you'd have a URL handy, now would you? Next time, I can check myself and not have to bother the list. Once we have the above info, Wikipedia tells us the Northwood chips were all 130nm, and that EM64T wasn't introduced until the 90nm series. The 2.4GHz/512KB/400MHz Prestonia CPUID 0F27 was released in April 2002. Nacona (90nm) with EM64T was introduced over two years later, in June 2004. Hmm. Well, it appears that CPUIDs are not unique. I've done some more research, and it appears that CPUID is just one more piece of evidence, which must be considered along with all the other known characteristics, in order to figure out what one has. One other piece of evidence that I needed, but did not tell the list, is that the chipset is the Intel 7500 chipset, which means that the effective bus speed is 400 MHz. (The actual bus speed is 100 MHz, but the processor operates at Quad bus speed, making the effective bus speed 400 MHz.) The CPU World web site which you pointed me to was very helpful though. After some more research, I believe that this is the right web page (watch out for a wrapped URL): http://www.cpu-world.com/CPUs/Xeon/Intel-Xeon%202.4%20GHz%20-%20RN80532KC056512%20%28BX80532KC2400D%20-%20BX80532KC2400DU%29.html The CPUID of 0F27h means that the S-spec number is either QML8, SL6EP, or SL6K2. QML8 is unlikely, since this is a Quality Sample. The bottom line: it is most definitely a 32-bit CPU. It does have 2 hyperthreads; so a Linux kernel with SMP enabled will see 2 logical processors. Now, my next step is to figure out what memory SIMMs to order. I'd like to install four 1G SIMMs, if they exist for this motherboard. But the devil is in the details. During POST, when the RAM is being tested, the following appears on the screen: SE7500CW20.86B.0034.P17.0522031027 Maybe that means something to you? -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/166902331.603426.1345689479917.javamail.r...@md01.wow.synacor.com
Re: [SOLVED] Is my processor 32-bit or 64-bit?
On Wed, 22 Aug 2012, Stephen Powell wrote: Hmm. Well, it appears that CPUIDs are not unique. I've done some more No, indeed they're not. But there is no 64-bit processor with CPUID 0F27h. Fortunately, there are precious few cpuids that are shared by 32 and 64-bit processors. The CPUID of 0F27h means that the S-spec number is either QML8, SL6EP, or SL6K2. QML8 is unlikely, since this is a Quality Sample. And you can find data about processors by their S-SPEC in http://ark.intel.com. The bottom line: it is most definitely a 32-bit CPU. It does have 2 hyperthreads; so a Linux kernel with SMP enabled will see 2 logical processors. Yes. And since it is netburst-based, you might actually be better off with hyperthreading disabled. Depends on what you'll use the box for. Now, my next step is to figure out what memory SIMMs to order. I'd like to install four 1G SIMMs, if they exist for this motherboard. But the devil is in the details. During POST, when the RAM is being tested, the following appears on the screen: SE7500CW20.86B.0034.P17.0522031027 Motherboard Intel SE7500CW2, BIOS 0034 P17. http://www.intel.com/support/motherboards/server/se7500cw2/sb/cs-006907.htm -- One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie. -- The Silicon Valley Tarot Henrique Holschuh -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120823031610.ga9...@khazad-dum.debian.net
[SOLVED] Is my processor 32-bit or 64-bit?
On Tue, 21 Aug 2012 04:18:52 -0400 (EDT), Stan Hoeppner wrote: Yep. CPUID 0F27 makes this CPU a Prestonia Xeon, 130nm, in essence a Northwood P4, the only difference being the model#, CPUID, and branding. Intel introduced EM64T (x86-64) with the 90nm chips. This CPU is 32bit x86 only. Stan, Stan, the hardware man! I knew you'd know! Where did you find the information that correlates CPUIDs with processor characteristics? I don't suppose you'd have a URL handy, now would you? Next time, I can check myself and not have to bother the list. Thanks to all who responded. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1357692593.580894.1345601046019.javamail.r...@md01.wow.synacor.com