• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

I DARE you to try to figure this crap out!

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
You know, the shorter you have had the system on, the more system resources you have. Windows handles resources horribly so a bench on the memory after having your system on for an extended period of time will have bad scores while a fresh rebooted system may score higher.
 
I just downloaded Sandra that runs on Win2K. I have a pcchips mb in my lab, it is a celeron 450 (overclocked 300A) 100mhz bus and 128mb stick of 60ns 168pin EDO memory. The mb is a M726 and it uses the ALI IDE controller also, I think it probably close to the same chipset that you have.
The memory benchmark comes out to 72/72.
I think for sure now I'd look at the memory stick. And check the bios to see what it thinks is in there.
 
One other factor that no one else seems to have touched on, the Quantum Bigfoot is a 3600 rpm hard drive and the slowest I have ever used. Absolute dog, speedwise. Even my wife hated it for publisher and E-Mail. I guess this system is the one that proves the statement that processor clock speed is not the most important factor when it comes to system performance.
Chuck
 
The ram is PC100 SDRAM. One is a PNY PC100 64MB CL2 SDRAM and the other is a Generic PC100 64MB CL3 SDRAM. They are currently running at CL3 cause it would not boot at CL2. It would freeze when it detected the hard drive. Anyway it is SDRAM. Both sticks work because I tested them in my athlon system and they both got around 300 something. I also took the ram out of my athlon system and stuck it in this system and ran just as slow. Maybe the motherboard is detecting them as EDO but I doubt it because Sandra detects them as SDRAM. Novice I think you are right about the hard drive. I didn't know it was only 3600 rpm. That would explain why it is going so slow. But I can live with the hard drive like that, the ram I can't.
 
I just put 32mb's of 7ns SDRAM in the lab pc.
At CAS 3: it was 150/150
at CAS 2: it was 160/170

So that's about double, but still sucks the big Kahuna.
Now it could be that the 32mb isnt enough due to W2k server eating it all up, and running Sandra may be causing memory swapping and falsely lowering the numbers, but I didnt notice any disk activity when the test was run.
I think that ALI chipset just sucks, if it was anything but my lab pc, I'd toss it.
 
You can't really even compare your scores on an athlon system with your scores on a celeron system! Notice the 83 mhz bus. Hence the reason for it to totally suck versus a 200 mhz bus score. I am no pro on scores for celeron ram so can anyone verify their memory scores to get somewhere near his so we can prove that the celeron sucks the big one.
 
Well the Athlon's bus is 200MHz DDR so really it 100MHz and the ram only goes 100MHz. And even if the ram went 200MHz it still wouldn't beat my ram that bad.(Well maybe I dunno) But there is something seriously wrong with this computer. I guess the chipset is just a piece of sh*t but I want to try everything before I say "screw this!!!". And take a hammer to the motherboard.
 
Well I guess there is NO reason why an athlon system ALWAYS scores better than a P3 or Celeron based system mainly because of this 200 mhz bus. The ram is going 100 mhz at ddr (which is 200 effectively). Although not a true 200 mhz bus, it is still going to score way higher than an 83mhz bus. I can't stress the point enough that different buses make huge changes in scores. Hence the reason why the scores are even lower when you run at 66mhz bus.
 
Well the reason an athlon system always scores better then a P3 or Celeron system is because they suck! The ram is not going at 100MHz DDR unless it is DDR ram. I know the bus speed makes a big difference on proformance but I doubt the ram is going that slow just because of the 83MHz bus. I think it is either the chipset or crappy drivers. Hopefully it just drivers because that I can fix. If it's the chipset then I'm screwed.
 
Oh yeah I can't believe I for to mention that the TNT2 is onboard and it set to reserve 16MB of ram. So really only 112MB of ram is usable. I can set it to reserve from 4MB-32MB in the BIOS. Now I doubt the lower amount of free ram is making it go that slow because MrChicken's 32MB ram proformed better then my 112MB ram. I think the fact that the ram is being reserved is what is screwing it up. I am going to put in my pci Voodoo 3 and disable the onboard video and see what happens.
 
Hmm, you did test the SAME ram in two DIFFERENT systems didn't you? You can't say the P3/Celeron system sucks just because you are totally biased toward AMD and you think your system is waaaay better than your father's. Come on, cpu speed has NOTHING to do with the MEMORY. Hello? Are you understanding me? Don't you get it that cpu speeds does not correlate with ram benches, BUT bus speed does! Sorry for all the caps and exclamation points but I really want to get the signal through that bus speed IS the reason for the scores being lower! Mr. Chicken's ram performed better because of the 100 mhz bus!
 
Then why is it that my K6-3+ on a 75MHz bus performs better then the ram on the Celeron with a 83MHz bus? I have used built at least 14 Intel systems and only about 6 AMD systems. So I am not biased towards AMD. Also my Athlon IS WAY BETTER then my dad's Celeron system. Hell the Duron 600 performs just as good if not better then the new Celeron 800 with the 100MHz bus. I am not a AMD zealot infact up until recently I only built Intel systems because they were BETTER. But now that, that is no longer true I build AMD systems. And if the rams performance has to do completely with the bus then why is it that the Athlon systems with NON DDR ram at a 100MHz bus beat the P3 systems with ram at a 133MHz bus?
 
Well, the answers as I see it to your question :

1) Memory bandwidth : The PCChips mobo is using a very odd chipset. Remember how VIA motherboards used to have lower memory performance (still do, a little bit, but not nearly as bad as before) than the Intel chipsets? Well, I totally suspect that's what you're seeing. Even including that, your bandwidth numbers are totally sucky. I don't know. Try reinstalling all drivers for the motherboard, Also, another BIG contributor is #2 below

2) It's totally crucial that your video is integrated. Usually an integrated video card shares memory with the system, which means that displaying anything on screen uses system memory bandwidth. You really notice this in a game like Q3, where your video card sucks up all the bandwidth it can get -- in your case, the video starves the CPU and vice versa. I don't know why the Voodoo3 doesn't improve the situation, but it should...
 
shawnmos

You mentioned that this computer was going to be used by your father for word processing... have you by any chance installed M$ Office ??? If you have than check in memory if OSA and FIND FAST are running. If these 2 programs are in memory than remove them and check if there is any difference in speed. I find these programs slow down processing by upto 25%.
 
Now that looks like what be near what my would be if I was running the test at 83mhz instead of 100mhz. If I get a chance tonight I'll put the memory back in and try running that pc at 83mhz and test it. If I get near the same numbers, you'll know it is just a sucky chipset.
 
OMG, you said those 3 dreaded words. ALI. might as well have been "piece of junk" those chipsets are A$$! altho i don't think it would make the entire computer bog down they pretty much are good for nothing. does the mobo even support DMA? how new is it? you can have the box checked in windoze of course and u may not even have hardware support for DMA. we all know how well windoze worx....

 
No 75 or 83mhz on my m726, I ran it at 66mhz and got 100/110. For laughs, and it was laugh, I tried the disk benchmark and got 1700. that's with a WD 2.5GB drive, PIO4 mode.
 
Your ram did just about the same as mine at a 66MHz bus. Shows you how much mine sucks. Oh well at least you hard drive did worse then mine. 😉
 
I knew that would make you feel better. After seeing how bad it sucked, I checked the ALI website and tried the Beta 2000 IDE driver, it's up to a whopping 2700 now. :-(
My other lab pc is Celeron 450 (300A oc'd) on a BX6 mb. It is so much faster than the pcchips board. It's mem numbers were about 260/300 with cas 3, and the disk gets ~3500 with a WD 1.6GB.
One thing this thread has really hammered home, pcchips mb's are only going in systems I give away. Well actually I did build a couple pcchips sytems that used the EX intel chipset and they ran pretty good, too bad they dont make them anymore. No built in vga, just sound.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top