Not Much Luck With LGA77 and DDR2

P4Man

Distinguished
Feb 6, 2004
2,305
0
19,780
did you guys see this article ?
<A HREF="http://www.tomshardware.com/motherboard/20040810/index.html" target="_new">http://www.tomshardware.com/motherboard/20040810/index.html</A>

thermtripping, clockthrotteling.. nice. Just when you need the power, *POOF* out goes the PC.

= The views stated herein are my personal views, and not necessarily the views of my wife. =
 

Mephistopheles

Distinguished
Feb 10, 2003
2,444
0
19,780
Hm, it could either be that:
1) Intel's asking for the impossible.
2) Mobo makers aren't doing their homework correctly.
3) A combination of 1 and 2.
4) Neither of the above.

Personally, I think it seems to be option 3, but there's only a small percentage of 2 in the 1/2 combination! Tom's even says "involuntary beta tester activity"... meaning the boards could have been more thoroughly tested... Of course, if it was easy to do what Intel wants, no mobo maker would have had any troubles... I mean, their processors are dissipating 120+W! So Intel isn't making it easy for them.

All I care, as a customer, is that I end up with a performance-rich, noise-poor computer. Let the mobo makers and intel fight each other, if need be.

(meanwhile, I'm probably getting an A64! :wink: )

<P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 08/11/04 02:44 PM.</EM></FONT></P>
 

Mephistopheles

Distinguished
Feb 10, 2003
2,444
0
19,780
Wow, can you imagine trying to run an eventual 4-way nocona-like processor at 3.6Ghz? <i>And designing the motherboard for <b>that</b>???</i> It's about 115W for each -> 460W typical. Because these processors can dissipate up to 150W peak, that means that the 4-way would give out 600W! Getting that into 2U units must be like trying to fit hell in a suitcase. Oh well... It's not a problem for them server buyers to spend another $100 for each super-ultra-cooler...

In contrast, a 4-way dothan would be only ~84W. Intel knows this, and is developing <A HREF="http://www.aceshardware.com/read_news.jsp?id=80000481" target="_new">a Xeon named "Whitefield"</A> that is exactly a 4-core P-M. (other indications of it being <A HREF="http://www.bayarea.net/~kins/AboutMe/CPUs.html" target="_new">dothan-based</A> exist throughout the web). A 4-core 2Ghz Dothan would dissipate not much more than those 84W now, let alone on the <b>65nm</b> process that Whitefield will be based on. Even a 4-way Whitefield system, if using 2Ghz Dothans, wouldn't come close to a 4-way nocona-like system.

Netburst is no good for this.

<P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 08/11/04 04:27 PM.</EM></FONT></P>
 

slvr_phoenix

Splendid
Dec 31, 2007
6,223
1
25,780
Taken from <A HREF="http://www.tomshardware.com/motherboard/20040810/intel_925_915-28.html" target="_new">http://www.tomshardware.com/motherboard/20040810/intel_925_915-28.html</A>:
While most of the motherboards demonstrated serious bugs during our test in July, <b>most of the problems have been solved by now</b>. We could not find any faults that were attributable to hardware design in the course of our test. Instead, problems such as wrongly set memory timings, incorrect temperature detection, premature shutdown at high load, performance deficits and other <b>annoying bugs can be solved through BIOS updates</b>, which all the tested manufacturers normally provide on their website at frequent intervals. ... Therefore, in the next few weeks, anyone reaching for a 915 or 925 motherboard (with the latest BIOS installed of course) <b>should no longer anticipate any serious problems</b> - as long as there is sufficient ventilation in the PC case and a good CPU cooler.<font color=blue>*</font color=blue>

<pre><font color=blue>*</font color=blue> = Bolding is my emphasis.</pre><p>
<pre><b><font color=red>"Build a man a fire and he's warm for the rest of the evening.
Set a man on fire and he's warm for the rest of his life." - Steve Taylor</font color=red></b></pre><p>
 

Mephistopheles

Distinguished
Feb 10, 2003
2,444
0
19,780
In general, it's not quite a wise idea to buy a brand-new technology. Problems are always there, be it Intel or AMD, because things are simply new! And grantsdale and Alderwood introduce loooots of new thing to go wrong...
 

slvr_phoenix

Splendid
Dec 31, 2007
6,223
1
25,780
In general, it's not quite a wise idea to buy a brand-new technology. Problems are always there, be it Intel or AMD, because things are simply new! And grantsdale and Alderwood introduce loooots of new thing to go wrong...
**ROFL** No flirking schiznit! Which is exactly why I'm going to wait a few months after WinXP SP2 is released, to judge how much of a headache installing it will be for me.

It's amazing how many people will spend a small fortune to jump onto the latest and greatest bandwagon with zeal and then complain about having problems as though anyone could expect everything brand new to be perfectly bug free. :O

Personally, I wait at <i>least</i> two months. ;) But then I also have an UPS with AVR, a larger power supply than I need, two low-RPM 120mm case fans for a not-OCed system, and a RAID1 redundant array. Minimizing annoying complications is generally a big factor for me. I like to spend time <i>using</i> my PC, not fixing it. He he he.

<pre><b><font color=red>"Build a man a fire and he's warm for the rest of the evening.
Set a man on fire and he's warm for the rest of his life." - Steve Taylor</font color=red></b></pre><p>
 

P4Man

Distinguished
Feb 6, 2004
2,305
0
19,780
>most of the problems have been solved by now.

Apparently, this says more about even worse problems that have been fixed.. I mean, a PC that just shuts down under stress, I can't see how that is a minor thing.

> annoying bugs can be solved through BIOS updates

Can a BIOS update reduce Prescott's powerdraw/heat perhaps ? Because that seems to be the real issue. I also have a hunch (no proof obviously) that maybe some BIOS's might be underreporting Prescot's temperatures as not too worry the user too much, but too hot is too hot, and it will shut down by itselve no matter what the BIOS reports, even if it only says 66°C.

>should no longer anticipate any serious problems

Well, according to Tom it isnt too bad then, when a PC shuts down while eg playing Doom3 ? I'd say that is more than a minor annoyance, and that problem apparently has not been fixed yet. Well, the problem isn't really available either yet (3.6 GHz cpu's) I guess, maybe that makes it okay :D

= The views stated herein are my personal views, and not necessarily the views of my wife. =
 

FUGGER

Distinguished
Dec 31, 2007
2,490
0
19,780
It takes a bunch of idiots to make a mountain out of a few grains of sand.

I am willing ot bet that they used the stock LGA heatsink without the full BTX case closed. The heatsink is designed to have the shroud and if that shroud is not present the heatsink will lose a lot of air CFM being pulled from outside the case.

There is no heat problem and machines are not shutting down all over the world. I am on LGA775 and running several different processors.

Most of the early problems have been fixed with bios updates but there are a few bugs to be worked out with each manufacturer. This is the same with all new product releases.

LGA775 is a better overclocking platform than socket 478 based upon my 3.4EE ES samples on both platforms.

With a BTX case and proper shroud around the retail heatsink the idle temp is around 38c with an ambiant of 78F

That shroud is very important part of the design, the fan is gimped without it. Lets see some tests with aftermarket heat sinks...

The thermal diodes are very accurate on the Asus and DFI 925X chipset mobo's and now go sub zero. The asus will drop to -36c and the DFI goes beyond -47c

<b>Kanavit owns you, get over it already!<b>
 

Cybercraig

Distinguished
Dec 31, 2007
2,058
0
19,780
You seem to like to disregard the obvious. Nobody in his right mind is going to jump on that platform right now. It reminds me of MSI sending 3 bad boards in a row out for review. The "build it and they will come" attitude exists only in fanboy minds and not in the real world. :smile:

Abit IS7 - 2.8C @ 3.5ghz - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - Yellowtail Merlot
 

FUGGER

Distinguished
Dec 31, 2007
2,490
0
19,780
Its a new platform, I dont expect many to jump on it. It needs some time to mature. No clue what the deal was with the MSI boards.

The new platform has many advantages over the previous chipsets.

I can pick stuff apart with every new mothboard release AMD or Intel. Stuff not labled, no voltage adjustments, missing settings for comming stuff and so on.

<b>Kanavit owns you, get over it already!<b>
 

trooper11

Distinguished
Feb 4, 2004
758
0
18,980
well im sure your aware that at least wiht this lg755 socket and chipset releases, intel is not wanitn got support overclocking and in fact actively fight to stop it, making bigger moves it seems wiht every platform.

but ill say this, i dont really blame them, i persoanlly dont care hwo far soemhting can overclock, that should never be a criteria as to wether any part like a cpu, motherbaord, or ram is a good product, unless of course its specifically labeled as to be used under those conditions. of course i know in the enthusiast community this is a bigger deal, but to the large majority of buyers that make intel and amd thier money, dont overclock and never will. and for me, i want to know whihc platform is best at the speeds it was meant to run at.

i think lg755 had to be done at osme point for intel, they would have to swallow a short term loos to get all these new techs into the market and saturate. it wont really come into its own for at least 6-9 months. ddr2 will have to be in greater supply and cheaper preices. and pci-e video cards wil have to be out across the board, form high end to entry level. once those things come together, it could be just as successful as any other intel chipset.


although there is the question of dual core and if it will be a new socket, forcing yet another change next year, that would not be good for lg755. but there is no word at all yet whihc way it will go.
 

endyen

Splendid
The big problem I have with socket T is the area of contact. From a current carrying perspective, trying to get a low voltage to induce current through a heated conductor, with a very limited contact surface, is just not a great idea. We are talking about a piece of copper at 60+c, the size of the point of a pin, in contact through pressure only.