RE: [H] New Intel 775 Pin Motherboards
775 has been out almost 9 months now, Chuck. It is not the base point of Intel's heat problem, which is incredibly well documented... yes, you can have a case built with enough fans to sound like a train that will cool down some of Intel's mini-furnaces they were shipping at the end runs of Prescott 478s, but the 775s are not a huge amount cooler. 775 was not designed as a cooling solution; Intel moved to BTX to try and do that, a radical requirement of new case types. And that's not saying BTX is a bad design, it's just saying it was warranted by Intel's heat concerns. I understand your anecdotal experience with Intel as having been positive.. and that's a good thing, and more credence to it. Intel's power consumption has been off the charts; that's not just some idle "some AMD backer says so" it's what Intel and others document: http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2353&p=4 The sheer reality is Socket 775 is dead. It is dead as it stands right now. Even Intel acknowledges that in the end, the Pentium IV did not turn out as they thought it would, with power consumption doubling, higher heat output and more constraints. Part of why AMD isn't putting the same constraints on people is that an idle temperature on the 3500+ is around 42C whereas an idle temperature on the Pentium IV 3.4G is somewhere around 55C. Which is not just a negative issue for the CPU, but the parts and components that sit right around the CPU. Here's the end result: Intel realizes the real promise is in the Pentium-M core. Which is why in the future, Intel is changing up and moving the direction of the Pentium-M, not a continuation of the disaster the Pentium IV has become. How that plays out will be interesting. CW
Re: [H] New Intel 775 Pin Motherboards
- Original Message - From: "Chris Reeves" <[EMAIL PROTECTED]> To: "'The Hardware List'" Sent: Sunday, September 18, 2005 10:53 AM Subject: RE: [H] New Intel 775 Pin Motherboards 775 has been out almost 9 months now, Chuck. It is not the base point of Intel's heat problem, which is incredibly well documented... yes, you can have a case built with enough fans to sound like a train that will cool down First, thanks a bunch. Your commentary is far easier to undersand than the professionally written technical articles in the many magazines etc. You better represent the average independent computer technician than somebody trying to twist their commentary for commercial gain. My business is very low volume. I have to keep it simple. I try to stock only 2 different models of cases and 2 different CPU's, the economy and the expensive. To add AMD to my line of CPU's I may have to stock 3 different motherboard models. Technology is moving so fast even keeping my stock very low profile, I am often selling obsolete components. Example: I should have moved to the 775's months ago. This may be impossible but I have to struggle to compete and limit my different models of motherboards to 2 and limit my different models of CPU's to 2. Intel's power consumption has been off the charts; that's not just some idle "some AMD backer says so" it's what Intel and others document: I power the Celerons with an Enlight 350 wattt Prescott rated case and the Pentium 4's with an Enlight 450 watt Prescott rated case, so power consumption is not an issue with me. The sheer reality is Socket 775 is dead. It is dead as it stands right now. Even Intel acknowledges that in the end, the Pentium IV did not turn out as Did I purchase the wrong P4 in the 775? I assumed there was little other choice in the P4's. I bought only one. Next time, if I want to stick with Intel should I purchase another P4 775 or something else? Even the Celeron went to 775 and I had to jump through a hoop to get the Celeron 478 for my Asus P4P800 SE motherboards. I did use the Asus P5P800 motherboard for the P4 775. I am waiting for the prices to come down before moving on up to the new video cards and DDR2 Memory. How that plays out will be interesting. As for now I have to just hang on and hope I buy the right Intel CPU's. I enjoyed the long period of Socket 478 Intel CPU's. Chuck
RE: [H] New Intel 775 Pin Motherboards
> First, thanks a bunch. Your commentary is far easier to > undersand than the > professionally written technical articles in the many > magazines etc. You > better represent the average independent computer technician > than somebody > trying to twist their commentary for commercial gain. > I would question the level of professionalism in some of the main stream technical online and dead tree publications :) But perhaps I'm just cinical
RE: [H] New Intel 775 Pin Motherboards
lated runtime for trueSpace at 4xAA. Given the task manager screenshot shown above, it is not surprising that trueSpace does not give up much in runtime even with another demanding application competing. Overall, the Athlon64 X2-4800 ties with the P4 840 EE, the 840D takes last place whereas the dual Opteron 252 sail to a comfortable win, despite the relatively poor performance in FineReader. - And this is in the traditional Intel dominated ground of 3D development apps, not gaming where Intel really gets hammered (I won't go into that, but you can hunt for the benchmarks) Heck, if you look at power consumption: http://www.lostcircuits.com/cpu/p4_840/18.shtml You see that they are talking about an Overclocked AMD running at 2.7Ghz (which would be about a 4800+ single core) consuming 40W, vs. a stock Intel 3.2 DC eating up... 141W? And this is not in high function.. at 68C? In the end, this is what I tell people: Intel has one of the best chips on the market, period.. without a doubt, a true technical marvel.. with the Intel Mobile line. But right now, I can't even hardly justify to anyone their "high end" CPU line. This may not be something you run into, and their Celeron line is very viable. Because right now, the best chipset solution going in either platform, AMD or Intel is the Nvidia Nforce chipset. And that chipset has the stamp of approval of both AMD/Intel who have showcased it at their own processor forums. It's solid, driver support is very solid, and compatibility with current shipping video cards is excellent. So, now you've got one solid chipset, Intel or AMD that you can really pick from. At that point, it makes it very hard for me to "push" anyone to Intel. We do build a bunch of intel boxes.. lots of people love them. If you were to hand me a stack of money, though, and tell me to spend it as I see fit to build a "good" box, it would be a rarity that I could, from a performance, budget, or standpoint of supporting under our 3 year warranty come up with an Intel box. CW -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Sunday, September 18, 2005 10:34 AM To: The Hardware List Subject: Re: [H] New Intel 775 Pin Motherboards - Original Message - From: "Chris Reeves" <[EMAIL PROTECTED]> To: "'The Hardware List'" Sent: Sunday, September 18, 2005 10:53 AM Subject: RE: [H] New Intel 775 Pin Motherboards > 775 has been out almost 9 months now, Chuck. It is not the base point of > Intel's heat problem, which is incredibly well documented... yes, you can > have a case built with enough fans to sound like a train that will cool > down First, thanks a bunch. Your commentary is far easier to undersand than the professionally written technical articles in the many magazines etc. You better represent the average independent computer technician than somebody trying to twist their commentary for commercial gain. My business is very low volume. I have to keep it simple. I try to stock only 2 different models of cases and 2 different CPU's, the economy and the expensive. To add AMD to my line of CPU's I may have to stock 3 different motherboard models. Technology is moving so fast even keeping my stock very low profile, I am often selling obsolete components. Example: I should have moved to the 775's months ago. This may be impossible but I have to struggle to compete and limit my different models of motherboards to 2 and limit my different models of CPU's to 2. > > Intel's power consumption has been off the charts; that's not just some > idle > "some AMD backer says so" it's what Intel and others document: > I power the Celerons with an Enlight 350 wattt Prescott rated case and the Pentium 4's with an Enlight 450 watt Prescott rated case, so power consumption is not an issue with me. > > The sheer reality is Socket 775 is dead. It is dead as it stands right > now. > Even Intel acknowledges that in the end, the Pentium IV did not turn out > as Did I purchase the wrong P4 in the 775? I assumed there was little other choice in the P4's. I bought only one. Next time, if I want to stick with Intel should I purchase another P4 775 or something else? Even the Celeron went to 775 and I had to jump through a hoop to get the Celeron 478 for my Asus P4P800 SE motherboards. I did use the Asus P5P800 motherboard for the P4 775. I am waiting for the prices to come down before moving on up to the new video cards and DDR2 Memory. > > How that plays out will be interesting. > As for now I have to just hang on and hope I buy the right Intel CPU's. I enjoyed the long period of Socket 478 Intel CPU's. Chuck
Re: [H] New Intel 775 Pin Motherboards
[EMAIL PROTECTED] wrote: Intel's power consumption has been off the charts; that's not just some idle "some AMD backer says so" it's what Intel and others document: I power the Celerons with an Enlight 350 wattt Prescott rated case and the Pentium 4's with an Enlight 450 watt Prescott rated case, so power consumption is not an issue with me. The primary implication is that Power Consumption = Heat. Requiring a stronger power supply is a secondary concern. Gary VanderMolen
Re: [H] New Intel 775 Pin Motherboards
sucks, would never build or recommend one. intel be damned. >:-} only bend pin I ever had I could not fix was with a boxed intel cpu that came apart in the sealed box and got trashed. intel did make it good, vendor basically said F U. better packaging would prevent this. BTW how the hell does changing the pin location change the temp of a CPU ? fp At 07:25 AM 9/18/2005, [EMAIL PROTECTED] Poked the stick with: What do yall think of this major change in design? Chuck -- Tallyho ! ]:8) Taglines below ! -- Call waiting, great if you have two friends.
RE: [H] New Intel 775 Pin Motherboards
It doesn't. It just diverts pressure on OEMs, as it makes it far easier to bend pins (which are now in the MB) during the installation phase. CW From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of FORC5 Sent: Sunday, September 18, 2005 2:49 PM To: [EMAIL PROTECTED]; The Hardware List Subject: Re: [H] New Intel 775 Pin Motherboards sucks, would never build or recommend one. intel be damned. >:-} only bend pin I ever had I could not fix was with a boxed intel cpu that came apart in the sealed box and got trashed. intel did make it good, vendor basically said F U. better packaging would prevent this. BTW how the hell does changing the pin location change the temp of a CPU ? fp At 07:25 AM 9/18/2005, [EMAIL PROTECTED] Poked the stick with: What do yall think of this major change in design? Chuck -- Tallyho ! ]:8) Taglines below ! -- Call waiting, great if you have two friends.
Re: [H] New Intel 775 Pin Motherboards
- Original Message - From: "Chris Reeves" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]>; "'The Hardware List'" Sent: Sunday, September 18, 2005 12:48 PM Subject: RE: [H] New Intel 775 Pin Motherboards Socket 775 is dead, in the longterm that is. In that saying, it is a very short life socket. Right now, it is your only choice. By early next year, it is expected to be replaced, making it one of the shorter lived sockets that I can remember. Nice answer there! Thanks. Quick socket changes is hell on trying to stock motherboard replacements for boards that go out in warranty. For those that go out outside of warranty, tough! I guess they will need a new CPU to go with their replacement motherboard. Maybe most suppliers will at least stock obsolete motherboards for a year after the socket changes for replacement boards. I can truthfully say that 90% of my motherboard replacements are due to the customer not following my advice to use a UPS. My customers even admit they had a huge bolt of lightning pop just as their computer went dead. I pride myself in being able to make any repair in less than 24 hours even if I have to replace the motherboard. Sudden changes in sockets may put an end to this. In regards to power supplies, if you use "good" PS, you're generally fine. But in practice, most people (those who just buy parts from you, not have a build done) tend to use a PS that comes with a case. A CPU that uses 200W alone means that high voltage video cards, etc. really weigh you down. I do not do over the counter sales. I do not even install new power supply boxes unless it is my Enlight case. I have seen many cases in my 8 years and never have I see a case as simple to work in as the Enlight midtower. When their power supply fails I move their computer to an Enlight case for the same price others charge to change out their power supply. I'm a high end buyer. I want to buy between these two options: What would you do if you wanted to build long lasting durable work horse computers that you could sell for around $950.00 and still make a decent profit and performance was secondary to endurance? Most of my customers are very happy with my Celeron 2.93 $700.00 computers as it meets their needs. I do not cater to the gamers who need high end computers. Intel 3.7G EE + a 7800GTX My first P4 775 was a 3.0 GHz. I was using 3.2 GHz and will ease back up to that area when the price comes down some. If you're using a PCI-Express Motherboard, and you're running a processor from Intel faster then 3.4G, then you're 450W is being used at almost 90-100% continuously.. with a single hard drive and a single CDROM drive... This is where your advice will especially come in handy, when I change to a video card and memory that require more power and still stay within the $950.00 price range for a Pentium 4 computer. Currently I am waiting for the price of the video card and DDR2 memory to come down. (though Intel specifically says it is not a supported design). So, you lower your power consumption by going with an AGP card, and even high end AGP cards like the 6800Ultra don't use the kind of power that a X850, 7800, etc. use. True, this is what is saving me now. It seems that Enlight is always slow to offer cases with the larger power supply that is needed. Currently Enlight should be offering cases with 550 or 600 watt power supplies. When Intel originally announced Socket-T (775) they had promised these speeds: 3.8Ghz, 4.0Ghz, 4.2Ghz, 4.4, 4.6, 4.8, 5.0, 5.06, 5.33Ghz. Next joke? feasible.. heat and power consumption issues made it to the point where Intel has been unable to get any yields on a 4.0Ghz CPU. Minimum performance for a work PC that surfs the Internet has already been far exceeded. The general public would be content with the Celeron 2.93 that I build with 512 MB of RAM. It is only a small section of the general public that is wanting performance that is not readily available in a Celeron 2.93 with 512 MB of RAM. True, the general public who has these are not satisfied with performance but it is due to the junk loaded and bugs etc., not the hardware specifications. Simply put, how much performance does it take to run things like Quicken, Peachtree Accounting, Microsoft Office applications etc.? Gamers are not even in my conversation. They are an entirely different breed with needs that are always increasing. Intel's system has some hitches with how this works.. so multi-core CPUs work effectively only with the Intel 955 or Intel 945 chipset. So, people Soon computer capabilities will exceed the capabilities of the user to multitask. My main need for a dual core CPU would be to handle resource intense tasks such as copying 100 GB from one drive to another while doing other tasks that require the user's attention
RE: [H] New Intel 775 Pin Motherboards
> Socket 775 is dead, in the longterm that is. In that saying, it is a very > short life socket. Right now, it is your only choice. By early next > year, > it is expected to be replaced, making it one of the shorter lived sockets > that I can remember. > Nice answer there! Thanks. Quick socket changes is hell on trying to stock motherboard replacements for boards that go out in warranty. For those that go out outside of warranty, tough! I guess they will need a new CPU to go with their replacement motherboard. Maybe most suppliers will at least stock obsolete motherboards for a year after the socket changes for replacement boards. I can truthfully say that 90% of my motherboard replacements are due to the customer not following my advice to use a UPS. My customers even admit they had a huge bolt of lightning pop just as their computer went dead. I pride myself in being able to make any repair in less than 24 hours even if I have to replace the motherboard. Sudden changes in sockets may put an end to this.< Or, will more likely mean that you will change boards & processors as well.. which is something Intel doesn't mind at all. >What would you do if you wanted to build long lasting durable work horse computers that you could sell for around $950.00 and still make a decent profit and performance was secondary to endurance? Most of my customers are very happy with my Celeron 2.93 $700.00 computers as it meets their needs. I do not cater to the gamers who need high end computers.< Ok, you want solid performance, low cost but great durability? At about $950? I wouldn't use a 2.93Ghz Celeron. Let's go a different way. Let's say I use a 3000+ AMD Venice Processor ($147) So, I'm spending $50 more then your Celeron 340.. but performance is better, and heat output is less. I'm not sure if you're using Integrated video boards or not.. let's say that you are.. ATI's RS480 is both durable and very well supported, it's not a screamer, but it's very functional.. several companies make those boards (Asus, MSI, etc.) Or, I can go a bit better, a Nforce4 board from Foxconn for $78 and a X300 card for $50. So, now I'm at $280 or so. 512MB of memory? $50. An NEC DVD Writer? $50. A 80G SATA HDD? $70. So now I'm less then $500. Add an Enlight case, maybe $70. I'd still make a $100 on a board that, unlike an Intel board for a CeleronD would upgrade and support PCI-Express, would support dual core processors and have a far greater "upside" in the longeterm to the end user. However, for a cheap solution, I'd use an RS480 board; so I'd be $147 + $70 + $50 + 50 +70 = $387. Add an OS? $470. So, at $700, I'd have a box that would be monumentally faster, single driver chipset support, and I'd pocket $230.. if you're selling Celerons at this price point, (which is great, good for your bottom line) this kind of design should go for more.. and better for the end user. My wife has been using a RS480 board for about a year.. it was a cheap build, and the performance is pretty good. >> > When Intel originally announced Socket-T (775) they had promised these > speeds: > > 3.8Ghz, 4.0Ghz, 4.2Ghz, 4.4, 4.6, 4.8, 5.0, 5.06, 5.33Ghz. > >Next joke? Well, at some point, it wasn't a joke.. Intel did propose those, and they noted them on their website as the plan for the future.. they just couldn't deliver any of it. >Minimum performance for a work PC that surfs the Internet has already been far exceeded. The general public would be content with the Celeron 2.93 that I build with 512 MB of RAM. It is only a small section of the general public that is wanting performance that is not readily available in a Celeron 2.93 with 512 MB of RAM. True, the general public who has these are not satisfied with performance but it is due to the junk loaded and bugs etc., not the hardware specifications. Simply put, how much performance does it take to run things like Quicken, Peachtree Accounting, Microsoft Office applications etc.? Gamers are not even in my conversation. They are an entirely different breed with needs that are always increasing.< I think you've got a shock to the system coming. :) As more and more people avail themselves of the internet, more and more online content becomes of a higher demand. Peachtree 2005 offers online training courses. Yes, part of it is bandwidth, but try doing the training courses at the same time as other things on a CeleronD which multitasks poorly, especially with graphic intensive issues. More and more products are doing things that (IMHO) may be fluff, but which clients are after. And it's getting more and more intensive as to what those things are.. we're not talking games, we're talking basic programs that you use. Look at the task tray on most computers that come in: Antivirus, Firewall, decoder software, anti-spam, and so on. The more tasks you need to do, the slower those tasks actually perform. Look at some of the "great" color inkjet printers th
Re: [H] New Intel 775 Pin Motherboards
- Original Message - From: "Chris Reeves" <[EMAIL PROTECTED]> To: "'The Hardware List'" Sent: Sunday, September 18, 2005 3:50 PM Subject: RE: [H] New Intel 775 Pin Motherboards It doesn't. It just diverts pressure on OEMs, as it makes it far easier to bend pins (which are now in the MB) during the installation phase. I think it is harder to bend pins that are on the motherboard. With the pins on the CPU it can easily get bumped or dropped on its way from the box to the socket. Also it is easy for your fingers to bend the pins as you are holding the CPU if you are not extremely careful and conscious of how close they are to the edges you are holding. The motherboard (and CPU) have plastic protective covers. You do not remove the pins cover from the motherboard until you are ready to install the CPU. Just as ZIF sockets means zero insertion force, you use no more than the force of gravity to pull the CPU down into a recess with 4 walls to align it and keep it in proper alignment. The CPU has a gold color triangle for proper orientation. I feel that if you professionally install the CPU with the same attention to detail that you install CPU's into ZIF sockets, it is actually easier. Do not touch the pins or any of the 775 contacts on the CPU. Avoiding touching the pins is like avoiding touching the ceiling in your house while just walking around the room. You would have to go out of your way to touch the pins. Chuck
Re: [H] New Intel 775 Pin Motherboards
I think it is harder to bend pins that are on the motherboard. With the pins on the CPU it can easily get bumped or dropped on its way from the box to the socket. Also it is easy for your fingers to bend the pins as you are holding the CPU if you are not extremely careful and conscious of how close they are to the edges you are holding. The motherboard (and CPU) have plastic protective covers. You do not remove the pins cover from the motherboard until you are ready to install the CPU. Just as ZIF sockets means zero insertion force, you use no more than the force of gravity to pull the CPU down into a recess with 4 walls to align it and keep it in proper alignment. The CPU has a gold color triangle for proper orientation. I feel that if you professionally install the CPU with the same attention to detail that you install CPU's into ZIF sockets, it is actually easier. Then you haven't used it enough. While it may seem as though the pins should be harder to bend, in reality, they bend extremely easily, and you WILL DESTROY your socket after a few insertions REGARDLESS of how much care you take. Read around the web--people that have installed thousands of processors are finding LGA775 sockets rarely live for more than a handful of insertion cycles. Greg
RE: [H] New Intel 775 Pin Motherboards
The only thing that this has going for it is that motherboards are generally cheaper then CPUs. Still don't like it thoughWhile on the subjectwas there a reason to stop using the slot designs? Seemed like a damn good idea at the time, although didn't last for long. Regards, Jason Tozer Database Analyst London Ext 1131 - 3SC.5 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Greg Sevart Sent: 19 September 2005 13:14 To: The Hardware List Subject: Re: [H] New Intel 775 Pin Motherboards > I think it is harder to bend pins that are on the motherboard. With the > pins on the CPU it can easily get bumped or dropped on its way from the > box to the socket. Also it is easy for your fingers to bend the pins as > you are holding the CPU if you are not extremely careful and conscious of > how close they are to the edges you are holding. The motherboard (and CPU) > have plastic protective covers. You do not remove the pins cover from the > motherboard until you are ready to install the CPU. Just as ZIF sockets > means zero insertion force, you use no more than the force of gravity to > pull the CPU down into a recess with 4 walls to align it and keep it in > proper alignment. The CPU has a gold color triangle for proper > orientation. I feel that if you professionally install the CPU with the > same attention to detail that you install CPU's into ZIF sockets, it is > actually easier. > Then you haven't used it enough. While it may seem as though the pins should be harder to bend, in reality, they bend extremely easily, and you WILL DESTROY your socket after a few insertions REGARDLESS of how much care you take. Read around the web--people that have installed thousands of processors are finding LGA775 sockets rarely live for more than a handful of insertion cycles. Greg *** This message and any attachment are confidential and may be privileged or otherwise protected from disclosure. If you are not the intended recipient, please telephone or email the sender and delete this message and any attachment from your system. If you are not the intended recipient you must not copy this message or attachment or disclose the contents to any other person. For further information about Clifford Chance please see our website at http://www.cliffordchance.com or refer to any Clifford Chance office.
Re: [H] New Intel 775 Pin Motherboards
The slot design was useful for only one thing--getting the cache closer to the CPU core so a faster bus could be use. Once cache was integrated into the die, the need for slots were out. Slots are... 1. more expensive to manufacture 2. more likely to have a connection failure 3. harder to develop adequate cooling solutions for today's hot processors 4. increase trace length dramatically, which limit bus frequencies Greg - Original Message - From: <[EMAIL PROTECTED]> To: Sent: Monday, September 19, 2005 7:23 AM Subject: RE: [H] New Intel 775 Pin Motherboards The only thing that this has going for it is that motherboards are generally cheaper then CPUs. Still don't like it thoughWhile on the subjectwas there a reason to stop using the slot designs? Seemed like a damn good idea at the time, although didn't last for long. Regards, Jason Tozer Database Analyst London Ext 1131 - 3SC.5 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Greg Sevart Sent: 19 September 2005 13:14 To: The Hardware List Subject: Re: [H] New Intel 775 Pin Motherboards I think it is harder to bend pins that are on the motherboard. With the pins on the CPU it can easily get bumped or dropped on its way from the box to the socket. Also it is easy for your fingers to bend the pins as you are holding the CPU if you are not extremely careful and conscious of how close they are to the edges you are holding. The motherboard (and CPU) have plastic protective covers. You do not remove the pins cover from the motherboard until you are ready to install the CPU. Just as ZIF sockets means zero insertion force, you use no more than the force of gravity to pull the CPU down into a recess with 4 walls to align it and keep it in proper alignment. The CPU has a gold color triangle for proper orientation. I feel that if you professionally install the CPU with the same attention to detail that you install CPU's into ZIF sockets, it is actually easier. Then you haven't used it enough. While it may seem as though the pins should be harder to bend, in reality, they bend extremely easily, and you WILL DESTROY your socket after a few insertions REGARDLESS of how much care you take. Read around the web--people that have installed thousands of processors are finding LGA775 sockets rarely live for more than a handful of insertion cycles. Greg *** This message and any attachment are confidential and may be privileged or otherwise protected from disclosure. If you are not the intended recipient, please telephone or email the sender and delete this message and any attachment from your system. If you are not the intended recipient you must not copy this message or attachment or disclose the contents to any other person. For further information about Clifford Chance please see our website at http://www.cliffordchance.com or refer to any Clifford Chance office.
Re: [H] New Intel 775 Pin Motherboards
From: "Greg Sevart" <[EMAIL PROTECTED]> Reply-To: The Hardware List To: "The Hardware List" Subject: Re: [H] New Intel 775 Pin Motherboards Date: Mon, 19 Sep 2005 07:25:10 -0500 The slot design was useful for only one thing--getting the cache closer to the CPU core so a faster bus could be use. Once cache was integrated into the die, the need for slots were out. Tidbit: The Pentium Pro actually had an integrated L2 cache (well, the cpu and cache were in the same package) and used socket 8. It was released some two years before any Slot-1 Pentium II was shipped.
Re: [H] New Intel 775 Pin Motherboards
- Original Message - From: "Greg Sevart" <[EMAIL PROTECTED]> To: "The Hardware List" Sent: Monday, September 19, 2005 8:14 AM Subject: Re: [H] New Intel 775 Pin Motherboards Then you haven't used it enough. While it may seem as though the pins should be harder to bend, in reality, they bend extremely easily, and you WILL DESTROY your socket after a few insertions REGARDLESS of how much care you Thanks for the tip. This tip will help many OEM's etc. cut back on the number of bare bones units they build up and other procedures that set up the scenario for having to change a CPU. With the pins on the CPU I still avoided as many CPU changes as possible. Most of my CPU changes were caused by changing motherboards, to using the pins more than once is a mute issue in these changes. If the life of the pins is 3 CPU installations, I will be ok. Over 90% of my units never experience one CPU change unless I am installing a new motherboard. It seems you agree that Intel really pulled a slick trick this time. Chuck
Re: [H] New Intel 775 Pin Motherboards
The slot design was useful for only one thing--getting the cache closer to the CPU core so a faster bus could be use. Once cache was integrated into the die, the need for slots were out. Tidbit: The Pentium Pro actually had an integrated L2 cache (well, the cpu and cache were in the same package) and used socket 8. It was released some two years before any Slot-1 Pentium II was shipped. True. However, the size of the die was too large to make it economical for anything but server usage. (die size = $$$) Plus, the Pentium Pro's cache, as you state, was not integrated into the core so much as it was slapped into the die package. Therefore, it couldn't achieve the same benefits of a huge bus width and low latency that true integrated cache (first on the Celeron A of all things...) brought. Greg
Re: [H] New Intel 775 Pin Motherboards
From: "Greg Sevart" <[EMAIL PROTECTED]> Reply-To: The Hardware List To: "The Hardware List" Subject: Re: [H] New Intel 775 Pin Motherboards Date: Mon, 19 Sep 2005 11:18:41 -0500 The slot design was useful for only one thing--getting the cache closer to the CPU core so a faster bus could be use. Once cache was integrated into the die, the need for slots were out. Tidbit: The Pentium Pro actually had an integrated L2 cache (well, the cpu and cache were in the same package) and used socket 8. It was released some two years before any Slot-1 Pentium II was shipped. True. However, the size of the die was too large to make it economical for anything but server usage. (die size = $$$) Plus, the Pentium Pro's cache, as you state, was not integrated into the core so much as it was slapped into the die package. Therefore, it couldn't achieve the same benefits of a huge bus width and low latency that true integrated cache (first on the Celeron A of all things...) brought. It was actually pretty close. The PPro clock for clock blew away the PII - the Celeron A and PIII with integrated cache was of course a bit faster, but not revolutionary. Those chips simple carried on where the PPro left off. The P2 was in many respects a step backwards, especially for servers back then. We talk about how the PPro was insanely expensive to manufacture and ironically the move to integrated cache with newer gen chips was not only for performance reasons, but for cost savings as well.
Re: [H] New Intel 775 Pin Motherboards
True. However, the size of the die was too large to make it economical for anything but server usage. (die size = $$$) Plus, the Pentium Pro's cache, as you state, was not integrated into the core so much as it was slapped into the die package. Therefore, it couldn't achieve the same benefits of a huge bus width and low latency that true integrated cache (first on the Celeron A of all things...) brought. It was actually pretty close. The PPro clock for clock blew away the PII - the Celeron A and PIII with integrated cache was of course a bit faster, but not revolutionary. Those chips simple carried on where the PPro left off. The P2 was in many respects a step backwards, especially for servers back then. We talk about how the PPro was insanely expensive to manufacture and ironically the move to integrated cache with newer gen chips was not only for performance reasons, but for cost savings as well. Yes, the PII without on-die or integrated cache suffered from an even smaller bus width and frequency to cache than the on-die but not integrated cache did. With process sizes as big as they were (450 and 350nm), it wasn't economical to put large amounts of cache in the processor die. However, with 250nm and smaller processes, the die size was shrunk enough to make it economically viable. If the die is reasonable, it is certainly a lot cheaper to integrated it into the core than to make a processor 'package' a la Slot 1. The PII was a definite step backwards with regard to the PPro. However, it was necessary to keep processor costs down. As I'm sure you recall, the PPro was marketed soley as a server chip. However, the PII was for desktops and workstations, and therefore had to fit in price points less than that of the PPro. With the gigantic die size of the PPro at that time, it simply wasn't possible. Ideally, Intel would have continued the PPro, and gave it a 350/250nm process shrink and maintained it as the server line, but clearly they didn't. :) Greg
Re: [H] New Intel 775 Pin Motherboards
It wasn't so much the die size overall as the PII had a larger die size overall (had a larger L2 cache then most PPros). It was that the large single die was harder to produce then 2 smaller dies. With the PPro if there was a problem with the CPU or the L2 cache both parts hard to be thrown out, but with the PII they could be tested individually so didn't have to throw out both. So the real reason was more to improve the yield more then decrease the cost of a CPU. - Original Message - True. However, the size of the die was too large to make it economical for anything but server usage. (die size = $$$) Plus, the Pentium Pro's cache, as you state, was not integrated into the core so much as it was slapped into the die package. Therefore, it couldn't achieve the same benefits of a huge bus width and low latency that true integrated cache (first on the Celeron A of all things...) brought. Greg
Re: [H] New Intel 775 Pin Motherboards
Uhh...unless I am just completely off base (and I don't think I am), Intel PURCHASED most of the cache that was placed on the PII Slot cartridges. Therefore, it wasn't an issue of yield improvements at all. Even then, the PII's used two 256KB cache chips, not a single 512KB chip. Greg - Original Message - From: "Eli Allen" <[EMAIL PROTECTED]> To: "The Hardware List" Sent: Monday, September 19, 2005 1:11 PM Subject: Re: [H] New Intel 775 Pin Motherboards It wasn't so much the die size overall as the PII had a larger die size overall (had a larger L2 cache then most PPros). It was that the large single die was harder to produce then 2 smaller dies. With the PPro if there was a problem with the CPU or the L2 cache both parts hard to be thrown out, but with the PII they could be tested individually so didn't have to throw out both. So the real reason was more to improve the yield more then decrease the cost of a CPU. - Original Message - True. However, the size of the die was too large to make it economical for anything but server usage. (die size = $$$) Plus, the Pentium Pro's cache, as you state, was not integrated into the core so much as it was slapped into the die package. Therefore, it couldn't achieve the same benefits of a huge bus width and low latency that true integrated cache (first on the Celeron A of all things...) brought. Greg
RE: [H] New Intel 775 Pin Motherboards
> Uhh...unless I am just completely off base (and I don't think > I am), Intel PURCHASED most of the cache that was placed on > the PII Slot cartridges. > Therefore, it wasn't an issue of yield improvements at all. > > Even then, the PII's used two 256KB cache chips, not a single > 512KB chip. > > Greg > Yeah I think that was the point, it meant that they COULD purchase the cache if they wanted which freed up production capacity for the core. This then drove down costs as they would have been able to get far more actual processors on a wafer than if it was integrated. Also at the time the packages that were being used in the industry were very expensive for larger dies compared to the ones we are using now.
Re: [H] New Intel 775 Pin Motherboards
Uhh...unless I am just completely off base (and I don't think I am), Intel PURCHASED most of the cache that was placed on the PII Slot cartridges. Therefore, it wasn't an issue of yield improvements at all. Even then, the PII's used two 256KB cache chips, not a single 512KB chip. Greg Yeah I think that was the point, it meant that they COULD purchase the cache if they wanted which freed up production capacity for the core. This then drove down costs as they would have been able to get far more actual processors on a wafer than if it was integrated. Also at the time the packages that were being used in the industry were very expensive for larger dies compared to the ones we are using now. It may very well be that the marginal cost for a larger die was greater then than it is now (largely due to the move to 12" wafers, I imagine), but let us not undervalue just how huge the PPro core was, even compared to today's dual core chips. The 350nm PPro with 512KB cache was 438 sq.mm, vs. the largest chips available today: the Pentium D dual core at 206 sq.mm and the X2 Toledo (2x1MB) at 199 sq.mm. The PPro was an extremely costly-to-manufacture beast, and still would be today. Greg
RE: [H] New Intel 775 Pin Motherboards
> > > > Yeah I think that was the point, it meant that they COULD > purchase the > > cache if they wanted which freed up production capacity for > the core. > > This then drove down costs as they would have been able to get far > > more actual processors on a wafer than if it was > integrated. Also at > > the time the packages that were being used in the industry > were very > > expensive for larger dies compared to the ones we are using now. > > > > It may very well be that the marginal cost for a larger die > was greater then > than it is now (largely due to the move to 12" wafers, I > imagine), but let > us not undervalue just how huge the PPro core was, even > compared to today's > dual core chips. > > The 350nm PPro with 512KB cache was 438 sq.mm, vs. the largest chips > available today: the Pentium D dual core at 206 sq.mm and the > X2 Toledo > (2x1MB) at 199 sq.mm. > > The PPro was an extremely costly-to-manufacture beast, and > still would be > today. > > Greg > Absolutely. I wish I still had some of my old copies of Byte magazine (it's a real shame they stopped printing it, I loved it). I remember when the PPro was the cover story, I was at school at the time and a few of us were huddles round Byte amazed at the images, specs and technical breakdown they did of it. :)
Re: [H] New Intel 775 Pin Motherboards
On 19 Sep 2005, at 17:48:130, Greg Sevart wrote: Yes, the PII without on-die or integrated cache suffered from an even smaller bus width and frequency to cache than the on-die but not integrated cache did. With process sizes as big as they were (450 and 350nm), it wasn't economical to put large amounts of cache in the processor die. However, with 250nm and smaller processes, the die size was shrunk enough to make it economically viable. If the die is reasonable, it is certainly a lot cheaper to integrated it into the core than to make a processor 'package' a la Slot 1. The PII was a definite step backwards with regard to the PPro. However, it was necessary to keep processor costs down. As I'm sure you recall, the PPro was marketed soley as a server chip. However, the PII was for desktops and workstations, and therefore had to fit in price points less than that of the PPro. With the gigantic die size of the PPro at that time, it simply wasn't possible. Ideally, Intel would have continued the PPro, and gave it a 350/250nm process shrink and maintained it as the server line, but clearly they didn't. :) Well, there was the... 333Mhz? Pentium Pro Overdrive, which IIRC was a P2 with 256KB of fullspeed cache in the package. :p -JB
Re: [H] New Intel 775 Pin Motherboards
On 18 Sep 2005, at 17:48:520, Chris Reeves wrote: Socket 775 is dead, in the longterm that is. In that saying, it is a very short life socket. Right now, it is your only choice. By early next year, it is expected to be replaced, making it one of the shorter lived sockets that I can remember. The Conroe samples I've seen thus far are S775, which suggests the socket is good into 2007 at least. Now, I see from what you're telling me you're buying AGP based Pentium IV 775 (so 865PE modified type boards). That's fine, lots of people use those (though Intel specifically says it is not a supported design). So, you lower your power consumption by going with an AGP card, and even high end AGP cards like the 6800Ultra don't use the kind of power that a X850, 7800, etc. use. http://www.techreport.com/reviews/2005q2/geforce-7800gtx/power.gif I'd say thats 'the kind of power' the X850XT and 7800GXT use. It's a little lower under load, but it's idle (non-3d) power draw is higher! -_-_ James Boswell [EMAIL PROTECTED] ICQ : 1653327 | AIM : TorazChryx MSN: [EMAIL PROTECTED]
Re: [H] New Intel 775 Pin Motherboards
From: James Boswell <[EMAIL PROTECTED]> Reply-To: The Hardware List To: The Hardware List Subject: Re: [H] New Intel 775 Pin Motherboards Date: Tue, 20 Sep 2005 06:30:07 +0100 On 19 Sep 2005, at 17:48:130, Greg Sevart wrote: Yes, the PII without on-die or integrated cache suffered from an even smaller bus width and frequency to cache than the on-die but not integrated cache did. With process sizes as big as they were (450 and 350nm), it wasn't economical to put large amounts of cache in the processor die. However, with 250nm and smaller processes, the die size was shrunk enough to make it economically viable. If the die is reasonable, it is certainly a lot cheaper to integrated it into the core than to make a processor 'package' a la Slot 1. The PII was a definite step backwards with regard to the PPro. However, it was necessary to keep processor costs down. As I'm sure you recall, the PPro was marketed soley as a server chip. However, the PII was for desktops and workstations, and therefore had to fit in price points less than that of the PPro. With the gigantic die size of the PPro at that time, it simply wasn't possible. Ideally, Intel would have continued the PPro, and gave it a 350/250nm process shrink and maintained it as the server line, but clearly they didn't. :) Well, there was the... 333Mhz? Pentium Pro Overdrive, which IIRC was a P2 with 256KB of fullspeed cache in the package. :p -JB There was also the Dixon - the PII laptop edition, topping at 366mhz, that had an integrated 256K of L2 on the die.
Re: [H] New Intel 775 Pin Motherboards
- Original Message - From: "James Boswell" <[EMAIL PROTECTED]> To: "The Hardware List" Sent: Tuesday, September 20, 2005 1:21 AM Subject: Re: [H] New Intel 775 Pin Motherboards The Conroe samples I've seen thus far are S775, which suggests the socket is good into 2007 at least. Let's hope so! I would hate to see a section of shelves with 20 different motherboard models on it and ask why so many. The OEM or computer shop would explain, "These are needed to replace any motherboard in any new computer sold in the last 3 years. These are Intel. AMD is on the opposite wall." The fewer models of motherboards we have to stock, the less loss we have by our product becoming obsolete before we can sell it. Chuck
RE: [H] New Intel 775 Pin Motherboards
James: You missed my point. Those cards you show are all PCI-Express. I was implying by design, AGP cards generally pull less then their PCI-Express cousins. :) -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of James Boswell Sent: Tuesday, September 20, 2005 12:21 AM To: The Hardware List Subject: Re: [H] New Intel 775 Pin Motherboards On 18 Sep 2005, at 17:48:520, Chris Reeves wrote: > Socket 775 is dead, in the longterm that is. In that saying, it is > a very > short life socket. Right now, it is your only choice. By early > next year, > it is expected to be replaced, making it one of the shorter lived > sockets > that I can remember. The Conroe samples I've seen thus far are S775, which suggests the socket is good into 2007 at least. > Now, I see from what you're telling me you're buying AGP based > Pentium IV > 775 (so 865PE modified type boards). That's fine, lots of people > use those > (though Intel specifically says it is not a supported design). So, > you > lower your power consumption by going with an AGP card, and even > high end > AGP cards like the 6800Ultra don't use the kind of power that a > X850, 7800, > etc. use. http://www.techreport.com/reviews/2005q2/geforce-7800gtx/power.gif I'd say thats 'the kind of power' the X850XT and 7800GXT use. It's a little lower under load, but it's idle (non-3d) power draw is higher! -_-_ James Boswell [EMAIL PROTECTED] ICQ : 1653327 | AIM : TorazChryx MSN: [EMAIL PROTECTED]
Re: [H] New Intel 775 Pin Motherboards
Which is why unless the OEM is building with those parts, they won't stock them just to be the kings of 24hr repairs & record profit losses. This is the 21st century after all, buy it when you need it. Out here in Jersey I think you'd find most of these hole-in-the-wall OEM's only stock what they build with and then a small sampling of CHEAP boards, sometimes one in the same. Oh, and did I mention (yet again) MARK UP, language barrier & dishonesty?? [EMAIL PROTECTED] wrote: Let's hope so! I would hate to see a section of shelves with 20 different motherboard models on it and ask why so many. The OEM or computer shop would explain, "These are needed to replace any motherboard in any new computer sold in the last 3 years. These are Intel. AMD is on the opposite wall." The fewer models of motherboards we have to stock, the less loss we have by our product becoming obsolete before we can sell it. Chuck
