Asus 16 phase vs. EVGA 10 phase

Rubycon

Weaksauce
Joined
May 1, 2006
Messages
125
Asus has a 16 phase VRM for LGA1366 whereas the Classified is 10 phase. Is 16 just overkill in this case as the latter board is supposed to be super stable when overclocked to "blood from turnip" levels!
 
Asus has a 16 phase VRM for LGA1366 whereas the Classified is 10 phase. Is 16 just overkill in this case as the latter board is supposed to be super stable when overclocked to "blood from turnip" levels!

Overkill is very subjective, and there's never overkill for extreme computing!!!

Give me 30 phase!!
 
It does add stability when overclocking.For the price they're charging for the Classified,you'd think they'd at least match features like that with the cheaper Asus boards.
 
It does add stability when overclocking.For the price they're charging for the Classified,you'd think they'd at least match features like that with the cheaper Asus boards.

Yes but look at the board around the CPU - the design is very different. (like the DFI boards)

I'm using a P6T6 Revolution now and the stability is fine but I really want an onboard 1394A and a PCI slot. The P6T6 Workstation meets these requirements and has PCI-X. The Classified omits the PCI-X bridge and SAS which I have no use for and generally is supposed to be the epitome of 1366 boards. Perhaps I should get both and keep the winner?
 
Asus has analogue phases whereas EVGA classified has digital phases - so in reality it's more efficient even though is has fewer of them. AFAIK, only DFI and EVGA classified offer digital PWMs on LGA1366 platform
 
Asus has analogue phases whereas EVGA classified has digital phases - so in reality it's more efficient even though is has fewer of them. AFAIK, only DFI and EVGA classified offer digital PWMs on LGA1366 platform

That's exactly the answer I was looking for! Thank you! :)
 
i am not sure about evga, but while dfi markets their boards a digital, they are actually "hybrid". not sure about this either, but the asus "16" phases pwm is not limited to the cpu, ie 2 or 3 for the ram, 1 or 2 for nb, etc. its all just marketing crap. i am sure there are ecs and biostar boards out there with 4 phase analog running just as overclocked, and just as stable as the asus or evga.
 
Asus has a 16 phase VRM for LGA1366 whereas the Classified is 10 phase. Is 16 just overkill in this case as the latter board is supposed to be super stable when overclocked to "blood from turnip" levels!

The number of phases are not always the most important. You also have to consider things like switching frequencies.
 
Biostar has 12 phase design. Asus p6t deluxe has 16 phases for CPU alone, memory is prolly 2 more
 
I personally think its a bunch of marketing gibberish... just like "Japanese capacitors" that are probably all made in Guangzhou, China. Japanese name... What exactly isn't made in China these days ? Digital vs. Analog... more gibberish..

Its all based on one-up-manship...

Having said that, I need to have both lol :)
 
Last edited:
Well having more phases digital or analog aside, the power delivery to the CPU is much more stable and the PWM it self runs a lot cooler.
 
Providing sufficient and clean power to the CPU can be accomplished by an excellent 4 phase supply design. Refinements and improvements over that would provide marginal OCing ability as far as the CPU is concerned with most of the benefits of more phases being in reliability (less power per phase per unit time allows smaller less expensive but higher precision devices - in general) and lower component cost in some cases (smaller filtering caps, inductors that have to handle less current - there are more of them but in general likely together cost less than fewer heavier duty devices).

All generalizations with plenty of exceptions to the above but what I am tying to get to is that at some point the number of phases in the voltage regulation circuitry, assuming decent designs is a diminishing return in terms of enhancing CPU overclocking. After all, if you supply the CPU with sufficient clean power it does not care how many, or how few, phases where used to create the power.

It is my understanding that what is critical is the "luck of the draw" on the quality of the on-board memory controller anyway when getting "blood" out of an i7.

In choosing a board I would be scouring the forums about how well the bios worked once a pool of good quality candidates was determined and over at another site it seems they like the D0 CPU stepping a lot if you have not aquired the CPU already.

My 2 cents (and likely all that it is worth anyway).
 
Yes it sounds as fishy as the X war with CD roms was years ago? Remember that one? CLV, PCAV, Tru-X, etc. :D

I will have the Classified in hand in about two weeks time, hopefully a bit sooner.

Does it make any difference whether the AUX 12V power has both connections strapped to the PSU?

I have an Ultra X3 1600W for reference.

A bit too high, my friends i7 920 D0 does 4.4 @ 1.240v only.

Is that with HT on or off? ;)
 
off topic, but

isnt it a fact that the only difference between c0 and d0 i7 920 is the serial number is printed 90 degrees offset, and the cpuid number has changed? how could this possibly account for better overclocking?
 
off topic, but

isnt it a fact that the only difference between c0 and d0 i7 920 is the serial number is printed 90 degrees offset, and the cpuid number has changed? how could this possibly account for better overclocking?

Afaik, its an improved stepping. Much like the G0 Q6600 over the regular Q6600
 
Intel has made note that it is only a cosmetic change, but obviously we can see that something else has changed too.
 
Yet again, I don't care for individual D0 results, the point was and still is that Classified works with D0s just dandy.

On my end it only needs 1.240 on the same motherboard,
 
Last edited by a moderator:
OK the last 3 posts where a bit off topic and lets play nice people, name calling is a infraction offense that can lead to a "vacation". Please re-read the forum rules if necessary.
 
Back
Top