Venice won't work with NF3/4 boards??!!

[xtrmeocr]

wow i didn't think the inq was this bad

*wonders if they (the inquirer) got this info from someone that had some serious user error(s) more then hardware error*...

more then likely they overheard parts of a conversation on something unrelated and tried to piece together a story. that seems to be their forte.

 

[Petr]

I would like to remind you that...

1. Palomino chips didn't work on some KT133 motherboards (it was not only Abit KT7A). Reason: hardware incompatibility.
2. Thoroughbred and Barton chips didn't work on some motherboards where Palomino worked. Reason: hardware incompatibility.
4. Winchester has stability issues on early s939 motherboards.

Those are pure HW issues, no BIOS bugs or MOSFETs not powerfull enough. The same may happen here. One old document mentioned AMD will implement P4's Thermal monitor like protection (slow down when overheating, not just catastrophic temperature shutdown) for 90nm parts. Winchester doesn't have such a feature. If they did this now and use a pin that was originally reserved, those motherboards that have this pin connected, may be useless (something like this was the case of Abit KT7A) because the CPU could refuse to work with those. We can't say now.

Inquirer sometimes has BS but often their source is quite reliable. They are more reliable than XbitLabs, for example.

 

[Sentential]

This is from X-bit labs:

Although the maximum thermal design power (TDP) for Socket 939 processors is set to 89W, like with Socket 754 CPUs, AMD demands a certain reserve from the mainboard makers, with a proposed TDP of 105W. It is expected that the heat dissipation of Socket 939 processors will grow up considerably with the launch of models on the 90nm tech process. Thus, it is quite possible that the future Athlon 64 will be as hot as the notorious Prescott-core Pentium 4.

Seems like if mb makers heeded AMD's recommendation then the new chips shouldn't be creating a problem due to power consumption.

AH HA...I WAS right. Thats not good for AMD users

 

[Sentential]

Well, I mean how much power could it draw?

The infamous C0 mobo killing Prescott was 115W at stock. These are 10W off. Thats not alot

 

[Petr]

AH HA...I WAS right. Thats not good for AMD users

That is not true. Venice has the same TDP as current chips. There is nothing like Prescott-like temperatures present. Also - see the 2.4 GHz dual-core running with 1.4V. Do you think they have thermal problems if they can put two 4000+ cores into single die without changing voltage?

 

[Sentential]

Also - see the 2.4 GHz dual-core running with 1.4V. Do you think they have thermal problems if they can put two 4000+ cores into single die without changing voltage?

It is obvious that we do not have the facts. However if the vcore is the same on Toledo that is generally a bad sign. In order to bring wattage down they would have to lower the vcore, not keep it the same. That would mean Toledo's power draw would be alot more if the vcore remains 1.4v

 

[xtrmeocr]

That is not true. Venice has the same TDP as current chips. There is nothing like Prescott-like temperatures present. Also - see the 2.4 GHz dual-core running with 1.4V. Do you think they have thermal problems if they can put two 4000+ cores into single die without changing voltage?

not to mention Intel TDP and AMD TDP is an apples to oranges comparison.

 

[Sentential]

not to mention Intel TDP and AMD TDP is an apples to oranges comparison.

Absolutly correct. I reserve all of my opinions on the matter until I get an official statement from DFI via Angry Games. THEY know whats going on more than INQ does. I refuse to speculate further...period

 

[TimoneX]

AH HA...I WAS right. Thats not good for AMD users

That was up at the launch of the winchester bro. I think alot of people feared AMD's move to 90nm would produce the same levels of heat Intel's did, but it just hasn't happened. I think AMD was just planning ahead and attempting to allow themselves some headroom. Naturally there'll be those board makers that chose to ignore AMD's recommendations and perhaps E0 cores won't be compatible with their products. Time will tell, but I don't think this is a power draw issue.

 

[Overclocker550]

arg this isnt good news. I guess amd screwed themselves. ah well looks like for those planning to go diego or venice, we may need to get a new revision of the same kind of mobo. revision A, revision B, revision C, etc. Whatever revision my mobo is, I may need to get a new revision that was made after Venice/diego so itll work. My mobo and all mobos were made before the new a64s so they had no clue what cpus to expect, they can only make mobos that work with cpus out at that time. I will still wait on word and see if a new bios ever comes out for my motherboard and if others with my mobo is having luck getting venice to work. Looks like alot of people will have to rma or ebay their mobos and get an updated one that works with venice. This is gonna be a pain but its the price to pay for using a wonderful venice

 

[man_utd]

Most of the inquirers information comes from rumors in the industry, or something on another site. I doubt they even have a venice to test with.

 

[Gautam]

Give the proc a chance at least...it hasn't even debuted yet.

 

[ronaldo]

http://www.xbitlabs.com/articles/editorial/display/cebit2005-2_2.html

While the dual-core products have 110W TDP compared to 105W TDP of modern single-core chips, AMD says the dual-core desktop AMD64 products will be able to operate on current mainboards fine.

 

[sleepyzman]

Gautam Give the proc a chance at least...it hasn't even debuted yet.

how true lets wait and find out

 

[@md0Cer]

The infamous C0 mobo killing Prescott was 115W at stock. These are 10W off. Thats not alot

That 105watt reserve for motherboard design also factors in the dual cored CPU's. I am pretty sure but not 100% on that one.

If the motherboards are designed with that reserve in mind, then Venice or San Diego (at least initially) most likely wouldn't be incompatible just because the board isn't able to adequetly provide power right?

Either way, aside from all this, I think it is wise to let time tell instead of wasting energy speculating.

 

[OC Detective]

Sigh, we already do know some facts from the sibling Opteron 252's . Firstly these run at a TDP of 92.6 (considering up to 2.4Ghz for CG steppings were in an envelope of 89W) you can expect little if any change to Venices up to 2.4GHz! Secondly people are already running the Opterons 252's in an Asus 940 with no issues after a bios update. Much ado about nothing this info from Inquirer - whilst they are generally a good source for information they are also sensationalist.

One other thing seems like the first 90nm cpus for retail with strained silicon (Opteron 252's) start from week 7......

 

[@md0Cer]

Sigh, we already do know some facts from the sibling Opteron 252's . Firstly these run at a TDP of 92.6 (considering up to 2.4Ghz for CG steppings were in an envelope of 89W) you can expect little if any change to Venices up to 2.4GHz! Secondly people are already running the Opterons 252's in an Asus 940 with no issues after a bios update. Much ado about nothing this info from Inquirer - whilst they are generally a good source for information they are also sensationalist.

One other thing seems like the first 90nm cpus for retail with strained silicon (Opteron 252's) start with from week 7......

Thats what I origionally thought.

I used to read the INQ, but they have me on a full BS alert with battle stations manned ever since I had to pull a bunch of articles from a website because of them.

 

[man_utd]

The INQ should be used as a rag for rumors, thats it. If venice turns out to be fine, then hey, they blew it again, who cares. However, I would think twice now before I step up. Would probably be safer to wait it out at the moment.

 

[OC Detective]

The thermal regulation protection probably refers to the maximum die temperature which if the Opteron 252 is anything to go by seems to be actually lower than previous steppings.

 

[rseven]

These stories do not always have a happy ending. A few years ago I bought an msi KT3 Ultra socket A mobo. It had the Via KT333 chipset and it was "supposed" to run at speeds up to 166fsb. Great, I put an xp1800 in it and figured when AND came out with 166fsb cpu I'd be in fat city. No such luck. The board was never able to run a Barton and there was no bios fix. While some folks got a Barton to run on there's I ended up having to buy a new mobo to upgrade. Stupid me bought another MSI board and I'm concerned that those MF's will screw me again. If that should happen you will see some posts that will curl your hair and make Sen's look very tame. (I assume you remember when Sen canned his MSI board)
We shall see.