Re: [Flexradio] more on CPU speeds, etc.

On 8/23/07, Bob McGwier [EMAIL PROTECTED] wrote: How many here have had working RTX-2000 development systems in their hands or really good development tools? Maybe as many as the Burroughs B5000. You can measure influence by volume or weight, I suppose. Without the B5000 (which Wirth

Re: [Flexradio] more on CPU speeds, etc.

At 11:42 AM 8/24/2007, Frank Brickle wrote: You can measure influence by volume or weight, I suppose. Without the B5000 (which Wirth bought in lots of five) people might still be thinking Algol was too complex to realize, so no Pascal, no C, no Simula, no C++, no SmallTalk, no Java, no C#... We

Re: [Flexradio] more on CPU speeds, etc.

I was taught that Lisp was the acronym for lots of insipid silly parentheses Jim - W4YXU - Original Message - From: k5nwa [EMAIL PROTECTED] To: Frank Brickle [EMAIL PROTECTED] Cc: flexradio@flex-radio.biz Sent: Friday, August 24, 2007 12:01 PM Subject: Re: [Flexradio] more on CPU

Re: [Flexradio] more on CPU speeds, etc.

I worked on the B5500 many years ago, and then went on to work on the B1700/B1800/B1900 series writing compilers for them! Anyone remember the KDF-9 one of the first, if not the first, stack machine designed to run Algol. Regards, John g0orx/n6lyt -- John Melton Sun IT CTO Office UK

Re: [Flexradio] more on CPU speeds, etc.

John Melton wrote: I worked on the B5500 many years ago, and then went on to work on the B1700/B1800/B1900 series writing compilers for them! Yeah, but that was *easy*, right? ;-) Actually a very slick, clever architecture. So many different kinds of groundbreaking ideas associated with

Re: [Flexradio] more on CPU speeds, etc.

I used to program a CDC-160 (ppt and Friden), IBM 1401 (cards) and Applied Dynamics AD/Four (analog). I hacked an interface from the 1401 to a DEC PDP 11/70 (RTL to TTL) so we could go from 556 cpi tape to the DEC's hard drive instead of cards. Saved some trees and lots of time. Does that

[Flexradio] more on CPU speeds, etc.

I note the following comment in the KB article: You definitely want to get a multi-core (AMD or Intel) or a Pentium D processor since future versions of PowerSDR will be taking advantage of the technology. Leaving aside the fact that today dual core processors aren't hugely expensive (in

Re: [Flexradio] more on CPU speeds, etc.

You definitely want to get a multi-core (AMD or Intel) or a Pentium D processor since future versions of PowerSDR will be taking advantage of the technology. Leaving aside the fact that today dual core processors aren't hugely expensive (in fact, one of the cheapest computers Dell

Re: [Flexradio] more on CPU speeds, etc.

I just upgraded my computer to a Q6600 configuration. The Q6600 (quad core) is only abt 70 bucks more than the E6600 (dual core). I honestly think any modern CPU (Athlon 3800 or greater, Pentium 3gHz, any dual core) will pass muster. I think video cards are also a moot point, other than avoiding

Re: [Flexradio] more on CPU speeds, etc.

Don't we already have a program to test the competence of the CPU? Ignoring the additional load of the firewire transfers, just running PowerSDR 1.10.1 demo mode with an I/Q file should give a pretty good idea of CPU utilization at the I/Q file's speed - no? I assume the firewire transfer load

Re: [Flexradio] more on CPU speeds, etc.

I have run the Flex-5000A with 1.91 beta and 1.10.1 software on the following PCs. HP dc7600, Pentium-D, 3.4 GHz, 1 GB RAM. HP dc7700, Core 2 Duo, 2.13 GHz, 2 GB RAM. Both worked fine, except for the sluggishness of rapidly slewing the tuning frequency. N9EWO will be testing it starting Thursday

Re: [Flexradio] more on CPU speeds, etc.

The reason stated above is just one of many reasons the KB article indicates that specifically defining a hardware configuration, whether it be in a configurator or a recommended hardware configuration matrix is a very subjective thing to quantify. I respectfully disagree. Flex knows

Re: [Flexradio] more on CPU speeds, etc.

At 01:50 PM 8/22/2007, Lyle Johnson wrote: The reason stated above is just one of many reasons the KB article indicates that specifically defining a hardware configuration, whether it be in a configurator or a recommended hardware configuration matrix is a very subjective thing to

Re: [Flexradio] more on CPU speeds, etc.

Rob Sherwood wrote: Probably comes down to RAM. If RAM starved, swapping to HD would likely really kill the program. Swapping is fatal for any pseudo-realtime program. Experience with the headless version of the SDR code (just DSP, no console) shows pretty conclusively that it's the

Re: [Flexradio] more on CPU speeds, etc.

Computer hardware is always ahead of software. Except when the hardware is tailored to the software, like the Lisp Machines, or Crays ;-) Or the Harris RTX-2000 series... Lyle KK7P ___ FlexRadio mailing list FlexRadio@flex-radio.biz

Re: [Flexradio] more on CPU speeds, etc.

Lyle Johnson wrote: Or the Harris RTX-2000 series... Good point, and it brings to mind the supreme example, the Burroughs B5000! 73 Frank AB2KT ___ FlexRadio mailing list FlexRadio@flex-radio.biz