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

[Sentential]

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.

I certinaly hope it does not come to that. I also PMed Skou to get a comment from Abit as well. Im thinking that the following manufacturers are going to be effected, juding by the quality of their power regs and previous Prescott problems:

Gigabyte (all)
MSI (all)
Asus (VIA based chipsets)
Chaintech (All)
Epox (Most)
ECS
Shuttle (all)

In short I wouldnt expect any issues on the following:

Asus (nVidia based chispets)
Epox (high end modles)
Abit (all)
DFI (all)

_________

DFI will probaly have the first BIOS solution followed by Abit then Epox. Unless you have one of these 3 I wouldnt be too optomistic.

Im pretty sure that DFI will be fine. For the most part they have the heaviest duty power regs. The evidence is the constant PSU problems cause by not running a 24pin PSU (indicated higher power draw). Abit is also included in this

 

[Petr]

Current motherboards should be prepared for 80 Amps. This is because dual-core A64 has preliminary specs calling for 80A (80A * 1.4V = 110W TDP). Current FX55 on 130nm draws up to 70A (104W TDP / 1.5V = 70A). On the other hand, Winchester has only 47A. This give a very big reserve for draw increase because of 2nd generation 90nm process. I don't see the point why should any motherboard have problems with supplying enough current.

 

[rseven]

I will wait and see what happens, but I have zero confidence in MSI. If they come up with a new bios that works I will be happily surprised. In the meantime I will plan on an upgrade for next year that will include a new mobo, cpu and video card. By then, what will and won't work will be known. Also, people will have a pretty good read on the nf4 boards and which is best. I know that there will be newer chipsets out by then but I prefer to stay one step behind the technology for reasons of price and bugs with the newest hardware.

 

[Overclocker550]

well it was mentioned that Venice takes less power than diego and dual core so maybe I can get away with a Venice on my gigabyte. Those looking for the higher clocked, 1mb cache diegos may need a new mobo

 

[situman]

damn i was happily thinkn about upgrading, now i gotta wait at least a month after venice comes out

 

[Sentential]

well it was mentioned that Venice takes less power than diego and dual core so maybe I can get away with a Venice on my gigabyte. Those looking for the higher clocked, 1mb cache diegos may need a new mobo

Im not worried about volts, like in life the AMPS are what's gonna kill ya. Winchester had a large ampage increase from Newcastle and the same is about to be reapeated here. I think its quite likely that what the INQ said is infact true. Most mobos can *barely* handle Winchester much Newcastle.

Nforce4 wont be affected much but Nforce3 is gonna be really REALLY sketchy

 

[Overclocker550]

amps as in currant draw? doesnt more volts increase amps? I could just leave the volts alone, ill only give up perhaps 100MHz but itll still crush my winny

 

[Sentential]

amps as in currant draw? doesnt more volts increase amps? I could just leave the volts alone, ill only give up perhaps 100MHz but itll still crush my winny

Im still very inclear on these matters so Im going to stop. I cant pull up the articles but its not the voltage amps or whatever its HOW the power is drawn. I really wish I could explain it better.

On paper Winchester doesnt draw much power but HOW it draws power puts alot of stress on the FETs

 

[deception``]

On paper Winchester doesnt draw much power but HOW it draws power puts alot of stress on the FETs

I thought we covered this on your Ultimate Winchester Thread....

MOSFET load is directly related to power consumption of the processor. What this means is simple:

Winchesters put LESS strain on MOSFETS than both 130nm Athlon 64's and Prescott systems.

deception``

 

[Super Nade]

A simple question:-

If AMD could get their 90nm CPU's to run cooler than their 130nm counterparts, what makes you think the next stepping, which is based on the same technology would require more power (run hotter?)?

 

[deception``]

A simple question:-

If AMD could get their 90nm CPU's to run cooler than their 130nm counterparts, what makes you think the next stepping, which is based on the same technology would require more power (run hotter?)?

It could be a lot of things in this capacity. Let's look at transistors for example. Chances are, AMD had to include a few more of these into the newer E cores, and they often add to the heat output of a processor. This is commonly seen with the Prescott core. Sure, it's based on a 90nm process, but those millions of transistors in such a small space account for its tremendous heat over the Northwoods. Of course, this is but a small sliver of possibilities, but I hope this helps

deception``

 

[man_utd]

Strained Silicon...

 

[Swatdog]

amps as in currant draw? doesnt more volts increase amps? I could just leave the volts alone, ill only give up perhaps 100MHz but itll still crush my winny

Naw, police tasers pack a punch of 50,000 volts with only 162mA (0.162A). You could be tased by that thing for 10+ hours and not have any side effects later.

 

[TimoneX]

1/10 of one amp can kill a human bean.

 

[I.M.O.G.]

A human bean? How does reproduction work for such a specimen?

 

[man_utd]

1/10 of one amp can kill a human bean.

depends on the voltage. You can have 1000000000000000000amps, but if you don't have enough volts to overcome the skin, it doesn''t matter.

 

[TimoneX]

It's generally expressed the other way, as is the case with lightning. .5 million volts and next to no current, but yes some voltage is required obviously.

 

[Sentential]

I thought we covered this on your Ultimate Winchester Thread....

MOSFET load is directly related to power consumption of the processor. What this means is simple:

Winchesters put LESS strain on MOSFETS than both 130nm Athlon 64's and Prescott systems.

deception``

Dont give a damn what the spec sheets say. Everyone having power issues with their boards (like I was) was running a Winchester. I have yet to see a newcastle provide the same 3d stuttering effect that a Winch does.

I cant explain it but it *does* put a higher strain on the FETs. I just dont have the technical know-how to explain it better than in those terms.

Take it for what it's worth, but I have not and will not back down on that issue. Something about the winchester core wreaks havok on current S939 boards (including DFI)

 

[deception``]

Dont give a damn what the spec sheets say. Everyone having power issues with their boards (like I was) was running a Winchester. I have yet to see a newcastle provide the same 3d stuttering effect that a Winch does.

I cant explain it but it *does* put a higher strain on the FETs. I just dont have the technical know-how to explain it better than in those terms.

Take it for what it's worth, but I have not and will not back down on that issue. Something about the winchester core wreaks havok on current S939 boards (including DFI)

In DFI's cases, they simply proposed higher power-requirements for their NF4 models. This is why there is the 24-pin stipulation. 24-pin units are not necessarily about providing more power; rather, it's all about a more even distribution of power. This way, individuals are able to successfully clock their s939 processors and run high-end cards in SLI (where applicable).

And Unfortunately, the numbers exist for a reason. One cannot just say that " I don't give a damn about the technical data," as it serves a purpose. I'm sure that you'd care to disagree with me, but this is not something that I am making up or even assuming. We all know that (1) Winchesters draw less power than those two and (2) power consumption is related to MOSFET load. Applying basic mathematics, 2 + 2 = 4, not 3.

Anyhow, the only way this could be possibly true would perhaps be due to the fact that the dual-channel memory controller of the s939 chips could produce a higher load, but that is a longshot within itself. More than anything, the likely cause of most (if not all) of the Winchester's problems is the infancy of the 90nm process, which cannot be ignored either.

deception``

 

[@md0Cer]

Could it be a function of the Winchesters requiring more amps? That could possibly be likely as they do draw less power by running on less voltage, but it is a good probability they could draw more amps and that in itself could be the strain on the fets?

Im not an expert in this stuff like deception here, but if I had to take a guess that would probably be it.