• 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.

EpoX 8K7A does *NOT* fully (nor officially) support Palomino...

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Insane3D, The week 51 NB must be part of my reaching the 160FSB on the A7V133, along with the RAM combo, and added copper HS on the ICS chip. (If anyone else does this and uses Arctic Silver thermal adhesive like I did, be sure to apply some clear silcone along the pins of the chip before hand) I have been through a few other A7V133's with different week NBs, and have not come close to matching this one.

I would be quite interested in your results with the Palomino at the high FSB of your IWill if you have the chance.
 
Is it possible that the motherboard manufacturers are not "offically" supporting the MP because AMD wants it to be geared strictly for multiprocessor systems?

 
Technonut wrote:

Pabster, I am just curious on how high you were running the FSB on the A7V133 when you were testing the Palomino? I had an 8K7A that I sold to replace it with a + version later since I like running my 4 drives as master on their own channels. My A7V133 has a week 51 NorthBridge that I have been running for a long while at an FSB of 158 with a mix of 2 128MB sticks of CAS2 Crucial, and 1 256MB stick of Infineon set at 3-2-2. For the last couple of days I have dropped down to 1440MHz with the FSB rock solid at 160 3-2-2 after I added a copper RAM sink to the ICS chip, and replaced the stock mounting pins that were alittle loose on the NB HSF with spring loaded pins from a Blorb that hold the NB HSF much tighter. It ran through the Prime95 Torture Test, looping 3DMark2001, and games without a hiccup."

A7V133 did not make it to 160, for me at least. 8K7A did it without breaking a sweat.

"I may hold off on the 8K7A+ for awhile if the performance is good with a Palomimo with the FSB at 160 on my A7V133. I know it will not have the boost that DDR has, but I am benching 655/738 in Sandra now, and it would give me a chance to run the Palomino and wait to see how other DDR boards will fare after they mature."

I received very similiar scores (630/720) with a regular Thunderbird 1.33 running at 1516 (168x9) on an Iwill KK266-R. Palomino needs the DDR architecture to shine. Certainly, if you can wait, there are new boards coming. nForce and SiS 735 both look promising. But at this very moment, 8K7A is hard to beat.

"EDIT: I know that I am reaching the limits of my 430 watt Enermax PS. I installed 4 extra 120mm fans to see what it would do, and the system rebooted it's self immediately after getting into Windows. I removed the extra fans, and it is back to normal. I have an Enermax 651 that I will be installing today to see if I can push my OC further, and also possibly be ready for some future MP action."

Yes, power (lots of it) is a definite requirement. I've pushed this Palo as far as possible with only an Enermax 365VP-E(FC), and I'll be trying to max it out with the Enermax 651VP-E next week sometime. It just can't handle the 2.2v core (drops quickly from 2.19 to 2.10-2.11 with a little load). I also notice the other rails being on the low side.
 
mchammer187 wrote:

"Pabster how are you testing your stability i have 2x256 MB of PC2100 and can only get 100% stability @ 141x10 FSB on a Tbird 1.2 Axia Y week 16 if i try to run it at a higher FSB then Prime95 craps out on me and gives me roundoff errors, have u tried Prime95 to test ur stability?"

Prime95 Torture Test, 3DMark2K1 loops, RC5/OGR cycles, SETI, and several others. It has passed with flying colors each time. This machine is 100% stable at 150x10 and 2.10v. I'm confident it'll be 100% stable at 1520 (160x9.5) after a power supply upgrade.

"Everything else works fine @ 147 FSB except Prime95"

Prime95 is the easiest test. 3DMark2K1 will also crash quite easily on an unstable overclock. I've had no trouble pushing high FSBs on this 8K7A.

"running
Epox 8k7a 8-8-8-2-2-2-2
512 crucial pc 2100 cl2.5
MSI GF 2 Pro
temps run @ 43C under load"

I'm using 8-8-7-2-2-2-2

Also 512MB Crucial PC2100, although I've also tried a couple Corsair PC2400 256MB modules for comparison. No real difference.

"trying to determine if it is the ram or crappy 400w psu"

There are an incredible amount of factors. First off, CPU. Each one is different, even from the same wafer. It is entirely possible that your chip hits the "brick wall" at that speed and won't go any faster, regardless of voltage. In my experience, the Crucial PC2100 performs and overclocks just as good as most Corsair PC2400 modules for far less $. I've had no memory problems with Crucial PC2100, even at 160 2-2-2. Another culprit could well be your power supply; not all are created equal, even of the same rating. I'm guessing the 400w is not an Enermax. I'd highly reccommend a 465VP-E or 651VP-E if you plan on seriously pushing the machine.

Have you tried raising your DDR voltage and/or CPU core voltage?
 
Shiva112:

"Is it possible that the motherboard manufacturers are not "offically" supporting the MP because AMD wants it to be geared strictly for multiprocessor systems?"

That is my thinking, precisely. 😀

 
"Is it possible that the motherboard manufacturers are not "offically" supporting the MP because AMD wants it to be geared strictly for multiprocessor systems? "

Yup, that's probably a big part of it. I know the site I bought my MP from, Monarch Computers, had a disclaimer on the MP product page that said.."For Use On The Tyan MP Motherboard Only"..so I have to believe that is the case.
 
thats odd, because from what I'm reading its the same exact chip in the desktop version.
Why not go for a simultaneous release? Intel already got the heads up on AMD by launching the P4 1.7 earlier than they anticipated.

In any case, what speeds are they going to debut the desktop Palomino on, and is it possible that it will come out before august?
 
Well, by all accounts a stock 1.33/266 Athlon =/+ a stock 1.7GHz P4. We could debate specifics forever, but by and large that is a fact. With the release of the 1.4/266, Thunderbird now outpaces a stock P4 1.7GHz machine (again, by and large, not 100% across the board.) I'm still amazed that when P4 vs Athlon comparisons are ran, it is always a 1.33 against a 1.7. How about we compare a 1.3 and a 1.4 Athlon/P4 to each other? I don't understand the logic in comparing a machine with a 400MHz clock advantage, except to show what I stated in the first couple sentences. 😀

The "official" desktop Palomino is rumored to be released initially at 1533MHz, with faster speeds after that. Originally slated for 1700MHz+ by the end of the year, supposedly 1600MHz is all we will see by Q4 of this year. I wouldn't count my chickens before they're hatched, of course. And Intel had better have Northwood ramping up 2GHz+ by that time, or I'm afraid the Palomino desktops are going to outpace them quite easily. (IMHO, Willamette, even at 2GHz would be toast to a 1533 Palomino.) This isn't meant to start a flame shoot -- this is a personal opinion and the reverse could certainly come true (a Northwood outpacing Palomino easily) but we'll just have to wait and see. It may well be another "no clear winner" comparison, but who knows.

As a side-note, am I the only one salivating at the idea of a Palomino 1533 or 1600 paired with PC2700 (DDR333, whatever you want to call it). Palomino is showing an amazing gain already over the original Thunderbird core even at 133, and at ~160 nearly equals PC600 RDRAM marks (without the high latency.)

 
You must log in or register to reply here.

TRENDING THREADS

Back
Top