• 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.

C533A Owners: Is Your Slotket Holding You Back from +897MHz?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Hey, I don't know about slotkets, but I can tell you week 14 566s of batch 478 and 532 suck! I'm 0 for 6 today so far! :Q
 
Maybe you can't reveal this in public, but what happens to all your rejects Wiz? RMA or do they go to non-OC system builders? Just curious.

I wonder if as Intel's CuMine voltage curve increases with faster P!!!s, the bin sorting becomes less rigorous, particularly since they're desperately trying to fill P!!! demand. If 850+ cores stable at 1.7v now qualify as within spec, then that takes a big chunk out of the cores that could have been C2s at 800+ at 1.65-1.7v. I'm still not clear on whether the Copperons are made in seperate batches from P!!! cores or not.
 
A bit OT, but this seems to be the post where all the Celeron2 guys are hanging out.

Does anyone know if the LN bios on the BH6 1.0x will run a Celeron2 and whether the jumpers on the ASUS s370-133 will be able to over-ride the BH6 bios settings? I would prefer not to have to flash my bios (makes too nervous) if I don't have to. 🙂

Napalm
 
It seems that most people with Copperons on Abit mobos are using the latest BIOS for their boards. If the BIOS you're using doesn't explicitly say it supports Coppermines or P!!!s up to 700+, you'll probably have to flash it. At boot it'll probably recognize the C2 as a "P!!!E", which is normal. For example my BX6r2 recognized the C533A as a "P!!!533E".

Depending on how much voltage you need, you may need to flash it again with default voltage set to "1.7v" to get 1.7 to 2.1 volts. I wouldn't depend on the slotket to regulate the voltage, as the BIOS always seems to override the slotket on my mobo.
 
Ok, I've been running tests on the BH6 with a MS-6905Master. Yes, I've been using the newset BIOS for the BH6 (R1.02), I updated the BIOS (QM) before I started the initial tests with the Asus slotket. This brought about some very interesting results... BTW, when I say stable I mean stable. The system is capable of running anything without any errors. I tested with Prime95 and Q3A for several hours each.

MSI MS-6905Master

850MHz (8.5x100) - 1.70v; Stable
876MHz (8.5x103) - 1.75v; Stable
952MHz (8.5x112) - 1.80v-1.90v; Will Post but is unstable, will load Win98 at 1.90v but crashes soon after... not temperature related (I'm using an Alpha PAL35 w/ silver compound).

Asus S370

850MHz (8.5x100) - 1.70v; Stable
876MHz (8.5x103) - 1.80v; Stable
952MHz (8.5x112) - 1.85v-1.90v; Will Post but is unstable, will not load Win98 at all.

So, we see that both slotkets seem to perform about the same (with a slight edge going to the MSI). I believe the MSI would be capable of running at 892 (at 1.80v or 1.85v) if the 105MHz FSB were available on the BH6. The Asus S370 would probably do well also but at a higher voltage (maybe 1.85v or 1.90v). At these voltages though very aggressive cooling would be a must, maybe with a peltier/water setup or just an awesome air cooler (high-end Alpha or GW). I'm not sure a GORB could handle it.

Abit BM6

850MHz (8.5x100) - 1.60v; Stable
876MHz (8.5x103) - 1.65v; Stable
892MHz (8.5x105) - 1.65v; Stable
952MHz (8.5x112) - Will test tomorrow when I have more time and can swap in better SDRAM.

Big difference, draw your own conclusions. But I'm leaning towards the BM6 being a more stable solution (with this CPU at least). Like anything else in elecrtonics when you shorten leads you get better signal. This very well may be the case.

I'll post results at 112MHz FSB on the BM6 tomorrow. Stay tuned! 🙂
 
Very good info there Jerry. I wish I had more time that I could spend with each CPU, but I just don't. I believe you are infact seeing the effects of a shorter route to the cpu, which translates to less electrical noise and overall better signal intregrity. Different motherboards do have their variables. As bad of a rap as Abit has gotten lately for their quality and piss poor customer service, some of the ol dogs are working out pretty good with these C2s. 🙂
 
The Slot 1 versions of the PIII Cumine chips have about a 0.05 V higher recommended spec I believe. Perhaps this is the slotket effect fudge factor.

For what it's worth, Abit Slotket !!! & ASUS P2B & Alpha PEP66:

533A @

800MHz (8x100) - 1.7; Stable
824MHz (8x103) - 1.7; Stable
840MHz (8x105) - 1.7; Stable
880MHz (8x110) - 1.8; Stable
896MHz (8x112) - 1.8; Not so stable - plays UT full swing for 20 minutes.
920MHz (8x115) - 1.8; Unstable - Photoshop in Win2000 for 10 minutes
960MHz (8x120) - 1.9; Crashes loading windows
992MHz (8x124) - 1.9; POSTs only
1064MHz (8x133) - 2.0; nothing

Note that my memory is generic PC100 Cas3 memory, and my motherboard doesn't do 0.05 V increments.

I'm getting a new motherboard, case, and PC133 memory next week. The new board is a Soyo 6VBA+133 (inexpensive). It's actually not going to be my primary board but it is an overclocking board so I'll test the CPU on it too if I get some time to waste. Doesn't directly apply to this thread though, since I have no other slotket to test.

By the way, you say S370 for the ASUS. Do you mean the old one? If so that is interesting, since the old one is not supposed to support FC-PGA. The current slotket is the S370-133.
 
After reading thru this thread and contrary to my earlier posting, I went out to a local computer store and bought an $8 generic slotket that the staff thought highly of, and replaced my Abit slotket (which I had bought for $10 at a computer fair) this evening. Low and behold, my 533a is running at 898MHz stable on my BH6 thru all benchmarks and games at a lower voltage, 1.75 vs the 1.8v required by the Abit. Amazing! What a suprise. This forum continues to amaze me, with the the quality info it delivers. My halo's off to you!
 
Okay, just to round out the thread here's my original post in table form:

C533A/Abit BX6r2 NW BIOS/128MB Corsair CAS2 PC100 SDRAM (set to 3-3-3 for 112MHz FSB and above)/Golden Orb & open case. AGP always set to 2/3. All slotkets have all jumpers set to their defaults.

Abit Slotket!!! (FCPGA-PPGA)

800MHz (8x100) - 1.65v; Stable
825MHz (8x103) - 1.65v; Stable
897MHz (8x112) - 1.85v-1.9v; Will load Win98 but not stable in any stressful apps.
936MHz (8x117) - 1.9v; POST only, no Win98 splash screen.
992MHz (8x124) - 1.9v; POST only, no Win98 splash screen.

MSI MS-6905 r1.1 (PPGA Only)

800MHz (8x100) - 1.65v; Stable
825MHz (8x103) - 1.65v; Stable
897MHz (8x112) - 1.85v; Stable
936MHz (8x117) - 1.9v; Will load Win98 but not stable. Probably just the limit of the proc.
992MHz (8x124) - 1.9v; POST only, no Win98 splash screen.

I'm convinced its a circuit noise issue, as Wiz points out. I imagine this is aggravated when a jumper is in the circuit, as required for FCPGA-PPGA slotkets like the Abit Slotket!!!.

Thanks all. This is a nice slotket round up. Anybody with an IWill II want to add their results?
 
One question KA, how did you get over 1.7v with the slotkets set to their defaults on the Abit board? Flash trick?
 
Yep. Set the default voltage to "1.7" and reflashed, giving 1.4v (I think) up to 2.1v. My mobo always overides ALL the jumper settings on both the Abit and MSI slotkets.
 
Well that's interesting. It probably makes no difference to me because of the limitations of my voltage settings, but this info just reaffirms my less than unbridled enthusiasm for Abit products.

Actually, in the several weeks at 840 since receiving my chip I have now had one BSOD, but I think that might have been related to the buggy scanner drivers I had installed. No problems since installing the newest update. Then again, one BSOD in weeks ain't bad because with Windows 95 BSODs were not an uncommon occurrence at all. No problems in Windows 2000, but I don't use much.
 
Unbelievable! Here's the update after I've gotten my hands on a 128MB stick of Micron CAS2 PC100 SDRAM.

Abit BM6

850MHz (8.5x100) - 1.60v; Stable
876MHz (8.5x103) - 1.65v; Stable
892MHz (8.5x105) - 1.65v; Stable
952MHz (8.5x112) - 1.70v; Stable so far after 2 hours of Prime95! At 1.65v it's stable in Win98, but gives me errors in Prime and kicks me out to the desktop in Q3 after about 10 min.

That cinches it! This is very cool... Seems like there are some pretty obvious differences in slotkets, your choice of slotket may even be more important than the choice of motherboard. But beyond that, a socketed board seems to be better than any slocket in terms of OCing. Really makes me wonder how far this 566a can go if I could just raise the voltage a bit more (1.75-1.85). Didn't Roger find a way to raise voltage allowances with the BM6? The "flash trick" just doesn't work on it, and I've tried it several times in just about any way you can imagine.
 
Actually, Thorn, what you've really proved here is maybe just that your memory is better. You can't say too much more about the Slotket until you compare the other ones with the same memory.

Anyways, 1.70 V 952 IS impressive. Even if isn't 100% stable, probably 1.75 V will be judging by your info. That said, too bad you don't have a 110 FSB setting. 935 MHz seems like it would be a sweet spot at 1.70 V. (As you can guess, running at max supposedly stable speed makes me nervous esp. since summer is coming and I don't have air conditioning.)

By the way, Wiz, you tested my 533A at 880 1.75 on a BX Master. That was with the Slotket !!! right?

Also, it seems to me that Prime95 is fine, but won't heavy duty gaming will crash a system at super high speeds faster than Prime95? Maybe it's because those vid cards generate so much heat, but then again that's the real world. With Prime95 all it does is stress the CPU and nothing else. It's easier to cool a case with no extra video generated heat.
 
That's what makes me curious about the C533As on a BM6. If you can get stability at roughly 0.1v less than when using a slotket, then a C533A @ 936 (8x117) at 1.7v would be pretty sweet.

Eug, you might want to consider running your case (and maybe yourself 😱) topless this summer, if you don't already.
 
Eug, that is a possibility. Both are now using high quality PC100 SDRAM (capable of 112MHz+). But the same chip is capable of 952 at 1.7v on the BM6 but can't do 952 at even 1.9v on the BH6 (regardless of which slotket I use). I know this isn't completely scientific... but it does cause me to raise an eyebrow. I'll try the same DIMM (from the BM6) in the BH6, run some tests, and see if it makes a difference. Ahhh... the troubles I go through in the name of science! 😉

BBL!
 
Ok, by using the same DIMM in both boards nothing changed. The 566 will not do 952 on the BH6 using either slotket while on the BM6 it's stable at 952 at 1.7v. Here's an interesting thing... MB reports the CPU is running on 1.73-1.74v, consistantly. That little bit of extra voltage may have a little something to do with it. I dunno.
 
Thorn:

My BH6 also always reports that it is running about 0.03v above what it is set to - so I don't think that is it.

Napalm
 
Thanks for all the time you guys have been putting into this. Killer Ape, Compuwiz, Thorn, Eug.... you've helped me decide that it's time to update my chip.

Low and behold, my current config is a 366 OC'd to 550 on a BM6 board.

If I understand this thread correctly, I won't need a slotket with this board. Get myself a Cel2 566A, and I'm off to the races.

Is this correct?
 
Thorn

Argh... damnit.. i didnt have to read that.. im in 850 but not 3dmark2000 stable, but 100% quake2 stable (go figure).. sigh.. and dmanit im using some noname converter.. it just reads 370c CPU card (then with little letters coppermine cpu card).. its so noname that there event aint a manufatr name on it...
Now ill have to go buy another slotket.. damnit..
I can get this beuty to post 950 too now, at 1.65V but it will not even consider loading windows.. using stock cooling.. more or less.

MSI u say .. hmmmm...
 
Cyruz: Have you tried anything higher than 1.65V @ 850? You say it posts at 950 @ 1.65v. It may not be the slotket, just that you need a little extra juice at 850to be stable, like 1.7v.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top