PCI Speed and Overclocking: A Closer Look at A64 and P4 Chipsets
by Wesley Fink on February 16, 2004 9:42 AM EST- Posted in
- CPUs
Sometimes things are not as they appear. Nowhere is this more true than the Computer Industry. Technology should be very straightforward, but all too often, marketing concerns get thrown into the mix and we end up with specifications with a marketing slant. It has been observed in reviews of the nForce3-150 boards that they often overclock better than VIA K8T800 boards. The nF3 sports a PCI/AGP adjustment in BIOS and it is usually stated that the nF3 should overclock better because of the fixed AGP/PCI bus.
Generally, we have also seen that Intel chipsets with Northwood processors reach incredible overclock levels. Intel has a widely published PCI/AGP lock in 865/875 chipsets, and it is clear that speeds like DDR550 (275x4 FSB) could never be reached without the fixed or locked AGP/PCI speeds. Most peripherals simply can't handle a PCI speed of 45.8 (275/6) or AGP of 91.7 (275/3) when specification is 33/66. Our Radeon 9800 PRO cards can rarely handle anything above about 70-72 on the AGP bus; this is typical of most current AGP cards.
Many tests have demonstrated the ability of Intel chipsets to lock the PCI/AGP bus, but questions are being raised again with the introduction of VIA's new PT880 chipset for the Pentium 4. VIA has been historically the only chipset maker not to utilize a PCI/AGP lock, but in the new PT880, VIA claims that they have implemented an asynchronous PCI/AGP bus. This is another way of saying that the processor and memory bus run at one speed and the PCI/AGP bus is not run at fixed divisors of that speed. This keeps the sensitive PCI and AGP cards in specification and allows a potentially higher overclock of the CPU and memory. Since this is VIA's first AGP/PCI, many are asking whether VIA really pulled this off - it has been such an issue with their chipsets in the recent past.
The most effective way to measure PCI/AGP bus speed is probably oscilloscope testing of bus input frequencies. To see an example of how this is done, you may want to check out tests on the nForce3-150 PCI bus at Lost Circuits. There is also a simpler tool available with the feature of measuring and displaying PCI bus speed, and that is PC Geiger. This device is a card that is mounted in a PCI slot and a digital display that provides all kinds of information - post codes, bus utilization, and PCI speed. The Taiwanese company, IOSS, markets PC Geiger in a joint venture with VICS in Japan. While PC Geiger provides many types of information, we will only be looking at PCI speed for these tests.
Facebook
Twitter
RSS
26 Comments
View All Comments
slewis5150 - Sunday, February 29, 2004 - link
How is this going to relate to pci eXpress?Pumpkinierre - Thursday, February 19, 2004 - link
So, Wesley#24, does this mean that some of the low maximum overclocks AT has shown for the A64 might have been as a result of PCI bus failure. Wouldnt it be possible to have a go at 234MHz for the Aopen and some of these other chipsets which supposedly had a PCI lock (nf3,SIS) but may be, like VIA, a dividor transition.Wesley Fink - Wednesday, February 18, 2004 - link
#23 - As we were testing in the article, none of the current Athlon64 chipsets have working PCI locks - and that includes nVidia and SiS (at least on the ECS). Some of the board makers do have a feature where a 1/7 multiplier kicks in at 233 and the PCI bus drops back to spec. We found that to be the case on the AOpen AK86-L which uses the VIA chipset.Jeff7181 - Wednesday, February 18, 2004 - link
Ummmm... I was pretty sure Sis and nVidia both had PCI/AGP locks and Via is the only one who can't get it right. Is this untrue?Xentropy - Tuesday, February 17, 2004 - link
Is this article going to be updated with the new information about 1/7 dividers and which boards have them? Not everyone reads the comments and would see the details.bigtoe33 - Tuesday, February 17, 2004 - link
Guys I bought the PCI geiger for Wesley from www.scan.co.uk They still show stock but I have just been informed its a discontinued product. If you want one I would grab one now.soki - Tuesday, February 17, 2004 - link
Good Job Wesley!Pumpkinierre - Monday, February 16, 2004 - link
Thank you Wesley!Wesley Fink - Monday, February 16, 2004 - link
As many have suggested as a possibility, it does appear different multipliers kick-in at 234 on the AOpen AK86-L. At a 233 setting PC Geiger shows PCI speed as 38.8, while at 234 the PCI readout drops to 33.2.TrogdorJW - Monday, February 16, 2004 - link
Is there any chance that a BIOS hack or something could adjust the PCI clock later in the boot process? You stated that you didn't load the OS and only booted into the BIOS. This seems a bit of a quick-and-dirty test, especially with the A-Open having acheived much higher results than other boards.My suggestion, in addition to including higher bus speeds, is to also get to the point where the OS is actually being loaded. You don't need to let it boot completely, but at least let it start. Get well beyond the BIOS POST test before assuming that the PCI speed being reported is "final".
On the other hand, the people questioning whether or not ALC655 and various other devices would run at 43 MHz... that's actually quite possible. Back in the day, I had a Pentium MMX 200 running at 250 on an A-Bit IT5H motherboard, which ended up being an 83 MHz bus and AGP and a 41.5 MHz PCI. (Ahhh... the good old Intel 430 HX chipset. Those were the days! /nostalgia.)
Later, I had Celeron 300A through 366 all running on 83.3 MHz bus, giving the same overclocked PCI and AGP. Finally, I am STILL running an A-bit BE6-II Rev. 2.0 motherboard on a 133 MHz bus. That motherboard has a 1/4 PCI divider, so PCI is in spec, but the AGP is running 2/3 and is at 88.9 MHz! I have a GeForce 4 MX 440 in it and it serves as my movie/music entertainment center.
Having been an overclocker since the days of the Pentium 166, I feel quite safe in saying that the "overclockability" of devices varies greatly. Some ATA/IDE setups can't handle more than 5 MHz out of spec (38 MHz or so), while others can clearly handle AGP and/or PCI speeds up to 33% out of spec. Only thorough testing will say for sure what any device can handle, and that's way beyond the ability for any one company/website to test.