Lies! 865/875 boards DO support Prescott!

[Crashman]

OK, so you've seen a bunch of AMD fanatics claiming crap about compatability, <A HREF="http://www6.tomshardware.com/motherboard/20030519/i875p-02.html" target="_new">here's an article written <b>TWO MONTHS AGO</b></A> that shows what this is all about:

Today, anyone entertaining the idea of getting a motherboard with the i875P chipset must ask themselves an important question: will the Northwood successor "Prescott" actually be supported by the board? It will, according to internal information from Intel papers, provided that an appropriate voltage regulator is used, in compliance with VRM specification 10.0. Add an adapted BIOS, and you should be good to go.

And the specification is even older than that! You see, Intel left it up to board manufacturers whether or not thy wanted to support the Prescott on their boards. Many have made claims that they have indeed provided their boards with the necessary hardware.

So what do we see at <A HREF="http://www.xbitlabs.com/news/mainboards/display/20030712011416.html" target="_new">X-Bit labs?</A>

In fact, mainboard makers have known that their current products do not support Prescott chips for weeks now, so, it is a mystery why, for example, ASUS still claims that their i865PE-based mainboard supports Prescott processor.

So why DOES Asus claim their boards support the Prescott? Perhaps it's because they actually FOLLOWED VRM 10.0?

So you see, you've all been fooled, but not by Intel, nor Asus. Instead you've been fooled by tabloid news that has been published not to inform you, but to scare you.

Who released the original story? Anandtech made a vague reference to an unnamed manufacturer who claimed this to be true. Perhaps this manufacturer was trying to delay your purchase until they could upgrade their own crappy boards? Or perhaps they were making more money off AMD products?

At any rate, I've held my tongue long enough, shame on you guys for repeating such garbage without investigating first. Shame on me for not looking up Tom's article sooner. I was kind of hoping some of you had read it, but I guess we don't read Tom's any more?



<font color=blue>Watts mean squat if you don't have quality!</font color=blue>

 

[Mephistopheles]

Excellent post there, Crashman. Excellent.

I saw the same evidence all around me: these are just rumors to scare us off Canterwood/Springdale. You did such a great job at posting this new thread that I might just *bump* this thing a gazillion times, just to be sure people got it:

<b>PRESCOTT IS SUPPORTED BY CURRENT MOBOS! </b>

BTW, <A HREF="http://www.hardocp.com" target="_new">HardOCP</A> was much more accurate: they said that "prescott support is entirely up to the manufacturer" - that is old news, but is right: Intel specced everything for prescott; all those silly mobo manufacturers have to do is follow specs! Like ASUS probably did.

<font color=red><b>M</b></font color=red>ephistopheles

 

[Crashman]

It's still possible for some motherboards to be manufactured improperly though. Look at Gigabyte's boards, did they repeat the redundant power or come up with new stuff? And what about Abit's 4-phase solution, doesn't Intel specify 3-phase power? Did they adhere to everything else and just give us a bit more, or...

Anyway, it seems obvious that at least SOME manufacturers adhered to the standard. Probably MOST of the good ones did. As soon as Intel sends me my engineering sample Prescott, I'll test it on all these boards and let you know!

J/K! I don't have these boards! But I do know of a reasonably good review site or two that does! Intel would go a long ways towards increasing consumer confidence by sending the review sites some samples!

<font color=blue>Watts mean squat if you don't have quality!</font color=blue>

 

[Mephistopheles]

It's still possible for some motherboards to be manufactured improperly though.

You're right, of course... Like I said, those silly mobo manufacturers could just adhere to the published specs...

<font color=red><b>M</b></font color=red>ephistopheles

 

[knewt]

Thanks for the info Crashman. Glad you're here to clear up this nonsense.

Wanted: Large breasted live-in housekeeper. Must be a good cook, organized, and willing to pick up after me.

 

[POPEGOLDX]

please post the link where you saw that they tested prescotts in 865/875 motherboards and they worked.

maybe some pics too.

would hate to be u later this year if you wrong....

making absolutes about unreleased hardware is foolhardy

 

[Crashman]

Notice: Sometimes unsubstantiated rumors ARE true! But the odds are greatly in my favor. If Asus had doubts, they would have pulled the compatabilitiy info from their site. And Asus would have a lot to loose if they didn't. So like I said, the odds are greatly in my favor.

<font color=blue>Watts mean squat if you don't have quality!</font color=blue>

 

[sirak]

Nice post, Crashman. The 865PE board I bought from ASUS claims a "new power design" to support the Prescott. Here is the <A HREF="http://usa.asus.com/prog/spec.asp?m=P4P800&langs=09" target="_new"> link </A>. It says it in the "CPU" section of the spec sheet.

 

[HardWareBoss]

Very good!! Time to wake up some people around here. Looks like a lot of people {AMD} Did not bother to read reports on the new Intel mobos on support for Prescott.

 

[HardWareBoss]

I stated last nite that it was an AMD ploy!!!

 

[longhorn2005]

Great post Crashman...

I agree with your point about ASUS..!!

It looks like AMD R&D Dept.("including fan-club") is now cooking plots to disgrace Intel..

Cheers~

 

[sabbath1]

Two words about the MSI 875P Neo motherboard: Stay Away!

Has anyone else experienced trouble with this board? While it probably would support Prescott, it's one of the worst motherboards I've ever had. Plagued with random game crashes, random broadband disconnections, boot crashes, unability to run certain memory, overclocking problems, MSI Live Update crashes, this mainboard is definitely NOT the one to get for any of you who wishes on a 865/875 board.

Bah, first I bought the P4P800, but exchanged it for this crap board because I wanted to be certain that PAT worked, and I actually got a board that actually crashed a few times BECAUSE of PAT! Hah, a Canterwood board that don't properly support PAT. Or maybe it was because of the overclocked voltage, but anyways PAT crashed a few times at boot.

Now I'm shooting for the ASUS P4C800-E Deluxe. Sure, damn expensive board, but I don't care anymore. Now I just want the damn thing to work.

So, two follow-up questions:

1) Has anyone else experienced these issues with the MSI 875P Neo? I use the LSR version.

2) How about the ASUS 865/875 series? Anyone with good/bad experiences?


My system: Intel Pentium 4 3.0, 800FSB / TwinMOS 1Gb DDR400 / MSI 875P Neo / Sapphire Radeon 9800 Pro / Antec True Power 550W / Western Digital Raptor / Hercules G.T XP /
Samsung DVD / Lite-On CDRW

 

[HardWareBoss]

I wonder why? Could it be that AMD is having major problems with A64? And it will be delayed again an again.

 

[Crashman]

Did you try slower memory timings? 865/875 chipsets put a lot of stress on memory, PAT adds even more stress, many modules have to be "underclocked" to slower than rated timings to get PAT to work.

<font color=blue>Watts mean squat if you don't have quality!</font color=blue>

 

[xeenrecoil]

Hey All;

It seems like memory manufactures were caught off guard by the i875 chipset and PAT even though they had plenty of forewarning by intel, i find it hard to believe intel just sprang it on them out of the blue, that would be bad buisness on intels part because it would set them up for failure.

So with that in mind, did Crucial, Samsung, and the other memory manufactures drop the ball?
Why are they still using outdated technology, they could make faster and much more stable modules and excede current specs with a smaller process.

Just off the top of my head i think they are using 0.15 micron process, i believe they just switched from 0.18 micron recently, we wonder why our memory bus has always lagged behind creating this large bottleneck in our system.

Well i believe that the memory buisness is fickle, yes even micron just want to make money, its pretty obvious by the way they do buisness that they dont really care about technology advancement but only about how fat their wallet is.

Intel is using a fabrication process far superior to Micron, you may ask yourself why this is the case, is Micron just THAT far behind intel...I dont think so! Micron just wants to make more money!

They are fully capable of using the .13 micron process and possibly even 0.09 micron process, they just dont want to period.

So what does that leave us with? Barely passable modules that can barely cut it on the i865, and i875 chipset. Unfortunately there is no one to kick them in the ass, and they have no desire to excellerate their product to perform beyond their current levels of performance.

Their behavior is sad and unchecked even in the event of damn near falling on their faces. You might say my memory works fine...does it really, does it excede specs? Does it perform on a level superior to ram in the last 2 years? They arent doing anything to improve their product, sure its faster in frequency but not in timings, and only companys that hand pick chips produce high performance modules, at a rediculous inflated cost rate to the consumer, and all of this is by accident as only a small percentage of chips can reach this high performance level.

The truth is, they are capable of pushing the technology much further then they are. Rambus is a primary example of a company that is hungry, just like a street fighter, they come out swinging they dont care who is in their way they just want to knock skulls, what drives them? well many things, poverty, environment, or perhaps they just want to be the best!

Rambus is already at 1Ghz and they arent stopping there, no infact they are planning on 2Ghz ram for 2004, now why cant micron make ram at these frequencies? well because they arent hungry and dont give a damn, they are the industry standard and are comfortable, they have no reason to excellerate their roadmap.

It was truly a sad day when Rambus was taken out of the loop by intel, this was largely in part to rambus higher cost which was handed down to the consumer, because poeple could get DDR much cheaper and go with a more cost effective CPU and yet still go almost as fast.

Rambus had a superior product but at an inflated cost to the consumer, thats why they failed, if they would have marketed their technology at a better price for the consumer everyone would be singing the praises of Rambus right now.

The truth is poeple are cost conscience, and micron and other large memory manufactures knew they could undercut rambus and not even sweat, it is obvious by the relaxed way they have introduced their products on their roadmap.

I hope Rambus kicks micron in the ass, they need a good swift kick, there needs to be a memory war...a really big one, a cut throat, im gonna kill you commie bastard type war, that damn near puts someone out of buisness, then poeple will wake up and smell the coffee.

Im going to stop now because i could go off for days about this.

XeeN

 

[Crashman]

Actually Intel said at first (way back) that they would be introducing the 800 bus with the Prescott and DDRII memory! They said regular DDR wouldn't clock that high reliably!

Then cheap enthusiast "proved them wrong" by pushing the technology. Refusing to buy RDRAM, looking for faster DDR, and they got DDR400.

DDR400 is...a problem, just like Intel said it would be. I read an article that explained why, unfortunately it was a little complicated for me to remember the details. It showed how DDRII overcame the limitations of DDR. At least, if the industry (JEDEC) follows Intel's recomendations. I don't even think there is any licenced technology involved, Intel just wants fast, stable computers for happy customers. Happy Intel customers that is.

Anyway, there are problems with the design of DDR that make it hard to produce stable stuff at both high frequencies and low latency. You get your choice...high frequencies OR low latency.

Right now the "fast, quick" stuff is comming out in low yields I think. I mean, if Corsair could produce DDR500 Cas2 and charge $300 a module...they probably would. And extreme performance enthusiast would probably cough it up!

As soon as one company figures out a way to make the modules better, others will probably copy that idea. The chips themselves shouldn't be a problem, as chips are available at even 2ns!

Of course it could just be that they need to redesign the PCB with shorter pathways using BGA chips, and such a redesign would not be profitable at this time.

<font color=blue>Watts mean squat if you don't have quality!</font color=blue>

 

[xeenrecoil]

heya crashman;

Personally i find it very sad that micron refuses to improve their product, i remember they were talking about going with QDR sooner, and for some reason they backed off their roadmap.

In my oppinion this was in favor of more profit, they are fully capable of producing superior products, they just simply are too comfortable. We both know that they are capable of producing a better chip with a smaller, more performance oriented process.

It just fustrates me that they dont care about the advancement of technology, they are supposedly the world leaders in memory technology, but at this point all they are doing is stiffling advancement, and the memory market is stale.

Memory technology is lagging behind and its all microns fault, they could have released DDR II much sooner and we would be seeing QDR for system memory application right now if they would have wanted to.

Some poeple might say Intel and AMD wasnt ready for the technology, well they are now, but where is it? Intel has been using QDR for two years now. sigh its so fustrating...and your right its not that they cant, its that it cuts into their profit margin too much.

XeeN

 

[sabbath1]

My Twinmos memory won't even run on CAS3, at least not on this motherboard. They can run on CAS2 though...and they're stable for quite a while. I've been running at SPD when running PAT though. Actually now I've ascertain that the mainboard crashes just as much without PAT so I guess the boot crashes must've been related to the weak overclocking capabilities of the board (this board runs at 203 FSB, and overclocking AGP, DIMM as well as CPU voltage by default.

What's strange with that my Winbond, rated at CAS2.5, would easily run stable at CAS2, Twinmos rated memory of CAS2, would not run at all in my system, not even with CAS2.5 or 3 with slowest possible timings. Death to this mainboard!

Also, like I mentioned, LiveUpdate crashes when trying to update things, only the BIOS part works (which is of course the most important part of LiveUpdate though), but still...

Any more problems? Probably


My system: Intel Pentium 4 3.0, 800FSB / TwinMOS 1Gb DDR400 / MSI *argh* / Sapphire Radeon 9800 Pro / Antec True Power 550W / 2x Western Digital Raptor / Hercules G.T XP /
SamsungDVD / Lite-On CDRW

 

[BrettJRob]

About the Prescott compatibility issue... can anyone with an ABIT IS7/IC7 board tell me if there is anything stated in their manual similar to ASUS's claim of compatibility?

 

[slvr_phoenix]

No offense Xeen, but if the RAM runs in single channel just fine, and the RAM runs in DC on nForce2 just fine, then one has to wonder if it is the RAM at fault, or if it is Intel's chipset that is at fault. It's easy to claim that the RAM manufacturers are to blame, but when the RAM runs just peachy-keen at low latencies in every mobo <i>except</i> an Intel 865/875 mobo...

Maybe even Intel screws up every once in a while.

"<i>Yeah, if you treat them like equals, it'll only encourage them to think they <b>ARE</b> your equals.</i>" - Thief from <A HREF="http://www.nuklearpower.com/daily.php?date=030603" target="_new">8-Bit Theater</A>

 

[eden]

One has to remember that Intel boards are NEVER meant for overclocking or tweaking purposes, so it is only foolish to even expect so much out of purchasing an Intel mobo. It's made for complete stability, and it does it right.

--
<A HREF="http://www.lochel.com/THGC/html/news.html" target="_new"><font color=red><b>Join the Tom's Hardware Guide Community Photo album, send us your pics!</font color=red></b></A>

 

[xeenrecoil]

Slvr_Phoenix;

no offense taken, you are intitled to your oppinion, its a valid possbibility.

I can tell you one thing though for sure, I am skipping the p4c and i875 and going strait to the Prescott, and DDR II, I want to avoid this mess entirely, i have read nothing but problem after problem with this latest CPU, Chipset combination, and it isnt worth it for me to take the chance, im fine with my P4G8X and p4b, untill the Release of the Prescott, and DDR II.

XeeN

 

[Mephistopheles]

Yes, indeed, that is a valid possibility...

ver vThere is always a disadvantage in going with the latest tech - you might run into these rough edges. Maybe when the 775-ball/semi-pin or whatever socket comes out and supports all Prescotts up to 5.5Ghz (*drool*) they'll have things straightened out a bit... DDR2 is on the safer side, anyway.

<font color=red><b>M</b></font color=red>ephistopheles

 

[Jake_Barnes]

Seems that AnandTech is "confirming" the report/rumor today.
<A HREF="http://www.anandtech.com/index.html#20040" target="_new">Link</A>.
I checked Asus site and they still claim Prescott support - anybody emailed them and asked them strightout? I checked a few others, Abit for one, and they do not "claim" Prescott support.

Seems just when I'm ready to dive in ... somebody moves the damn pool.


<b><font color=red>It depends on what the meaning of the word "is" is. - Bill Clinton<b><font color=red>

 

[Mephistopheles]

Anandtech merely quoted old news from Xbitlabs. There is no new info at all there.

<font color=red><b>M</b></font color=red>ephistopheles