PCIe And CrossFire Scaling: Does Nvidia's NF200 Fix P55?

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
[citation][nom]jtt283[/nom]Probably tongue in cheek, but relevant; can the traces of these mobos handle three cards that don't use PCIE power connectors?[/citation]

No PCIe connector? Its not even worth testing. Why would you buy 3 low end cards?
 
The MSI and EVGA boards used in this test/article retail any where between $329 - $365...there are many X58 boards that are much cheaper and offer the same features, functionality, and overclock just as well if not better. And, on top of that the NF200 chip takes up some overhead, so if you do use a single gpu in these mobo's you actually realize about a 2% performance hit. These mobos are only good for multi-gpu set-ups, and at that price, I find it hard to justify the cost over similar performing and cheaper X58 mobos.

And, the argument that the new i5 chips perform better at games that the Skt1366 i7 920 fail to realize that the i7 920 can run any game without issue at stock speed let alone when you push it to 4GHz plus.

These P55 boards with the NF200 chip are a nice experiment but I really don't see the advantage over an X58 set-up.
 
Just another example how awesome the p55 platform is, something for everyone. A i7 860 or even i5 750 can be part of your family whether its in a micro chassis htpc or the ultimate water cooled gaming rig.
The nf200 is not a exotic chip, its a very inexpensive logic switch thats copyrighted by Nvidia. I've read it costs the m/b makers 4 dollars.
 
The nf200 is not a exotic chip, its a very inexpensive logic switch thats copyrighted by Nvidia. I've read it costs the m/b makers 4 dollars.
Click to expand...

:lol: :lol: :lol:

And you seriously believed that?

From what I hear the NF200 chip is around $30 bulk.
On top of the chip costs you have to worry about a more complex layout on the motherboard, higher energy drain (more or better voltage regulation) and the fact that the NF200 chip is big and HOT (more elaborate cooling setup and higher temps).
Just as pure speculation, I would guess it costs the manufacturer at least $50 to implement this chip.
 
Two wrong statements above.
There is a separate fee for sli certification, and that also puts a sli bridge in each m/b box. Every sli certified board does not have this IC.
Crossfire certification is given to any pci-e slot even 4x as shown by the GIGABYTE GA-P55M-UD2 LGA, thats a clue on where Ati stands on these things.
 
How are they wrong again?

Two separate things.

1) $5 charge to make a x58/p55's BIOS SLI certified.

2) NV200 costs a hell of a lot more than $4 and makes the board more complex/expensive.

Care to tell me what is wrong with those statements?
 
This review is not really objective.

There are important things left out. One thing the author hides quite well is he's using dual channel memory for the 1366 platform (based on the configuration, unless he left something out), instead of triple channel. Stating this makes no difference, against what other benchmarks show isn't convincing. Then he states LGA1156 overclock better, whereas it's become common knowledge they do not overclock as high, and they require more voltage at the same clock speed.

Normally, you write a supposition with supporting facts. We get supposition followed by supposition.

A few other things. This N200 slows down the motherboard when using one card in almost all situations. Not a big deal if you are really going to be using more than one all the time, and it's not a lot, but it's worth mentioning.

It does add power use and cost. Why wasn't a power use test done? Obviously, this is still an inferior setup to x58, but if the power use and cost is a lot less, maybe it's good for some people. But, neither data points were included. But probably once you add this thing in, both power use and cost go up quite a bit.

Again, keeping mind the incomplete data, the suppositions without supporting facts, and the skewed benchmarking where the x58 was run in dual channel mode instead of triple channel (which is how most people will run these systems, obviously), this review does not convince, although it does show the N200 can be effective. But at what costs?
 
[citation][nom]chunkymonster[/nom]The MSI and EVGA boards used in this test/article retail any where between $329 - $365...there are many X58 boards that are much cheaper and offer the same features, functionality, and overclock just as well if not better.[/citation]


I completely agree. The only situation in which this may be useful is if you already have the CPU (LGA1156), and you decide then that you want to have 3+ graphics cards. Even then, you can probably sell the CPU (And anything else platform specific), buy an i7 920 + X58 and still come out on top.
 
[citation][nom]ta152h[/nom]Then he states LGA1156 overclock better, whereas it's become common knowledge they do not overclock as high, and they require more voltage at the same clock speed.[/citation]

When you lie once in a conversation, everything else you say becomes useless blather. We've already had this conversation on another thread, so you're fully award that what you've said is blatently untrue.

If you want a discussion, just leave the part where you lie...out of it.
 
All I can see the X58-with-nf200 mobo being necessarily beneficial for is the gamer who also does super-computing applications on his home PC. This would offer the flexibility of 3x GTX 295's running with extra bandwidth, and then, when not calculating the the 1,000-Sagan-th (look it up) digit of pie, they can play Crysis at "Billions and Billions" of FPS.
 
[citation][nom]mfarrukh[/nom]Just Give us the Dman FERMI already[/citation]
No, you cannot see my FERMI. Unless you're a hot chick, then we can discuss this over dinner.
 
[citation][nom]2shea[/nom]The only games available that can really max out anything are crysis, WIC, total commander and that's pretty much all.For the rest pretty good article though[/citation]

You forgot STALKER, right up there with Crysis for killing a vid card.
 
[citation][nom]techguy378[/nom]Nobody needs more than two graphics cards in a computer for gaming.[/citation]

Hehe, I love sweeping generalizations!

"No one will need more than 637KB of memory for a personal computer. 640KB ought to be enough for anybody." --Bill Gates
 
[citation][nom]falchard[/nom]Anyone else thinks its odd the days of tech we live in? An Intel Board running 3 AMD video cards crossfired better using a nVidia chipset.[/citation]
keep your enemies closer?
 
Completely agree with others. Why would you settle for P55 instead of x58 while running 3 GPUs altogether. Nice try though it defeats the purpose of having high-end components for the build.
 
[citation][nom]ta152h[/nom]This review is not really objective. There are important things left out. One thing the author hides quite well is he's using dual channel memory for the 1366 platform (based on the configuration, unless he left something out), instead of triple channel. Stating this makes no difference, against what other benchmarks show isn't convincing. Then he states LGA1156 overclock better, whereas it's become common knowledge they do not overclock as high, and they require more voltage at the same clock speed.Normally, you write a supposition with supporting facts. We get supposition followed by supposition. A few other things. This N200 slows down the motherboard when using one card in almost all situations. Not a big deal if you are really going to be using more than one all the time, and it's not a lot, but it's worth mentioning.It does add power use and cost. Why wasn't a power use test done? Obviously, this is still an inferior setup to x58, but if the power use and cost is a lot less, maybe it's good for some people. But, neither data points were included. But probably once you add this thing in, both power use and cost go up quite a bit.Again, keeping mind the incomplete data, the suppositions without supporting facts, and the skewed benchmarking where the x58 was run in dual channel mode instead of triple channel (which is how most people will run these systems, obviously), this review does not convince, although it does show the N200 can be effective. But at what costs?[/citation]

This is incorrect. Both LGA1366 and LGA1156 can achieve 4.2GHz on average chips, although more voltage required on LGA1156. BTW, I never heard someone who want to achieve 4.2GHz on LGA1156 but failed so far.
(p.s. HT needs to be turned off at 4.2GHz for both LGA1156 and LGA1366 i7)

Furthermore, not much voltage increase being required on LGA1366 for aggressive OC is due to its worse BJTs which has a lot of leakage current even when they are in the idle state. Since you have a lot of extra unnecessary current for LGA1366 (and therefore LGA1366 runs much hotter which is not good), you don't need to increase much voltage to raise the power for high OC according to P=VI.

Read the article from the link below. It is about the new BJT & PCU(power control unit) on LGA1156 platform which acts as a TRUE switch. This means that there's almost no leakage current at idle state.

http://www.anandtech.com/cpuchipsets...spx?i=3634&p=4
 
[citation][nom]andy5174[/nom]This is incorrect. Both LGA1366 and LGA1156 can achieve 4.2GHz on average chips, although more voltage required on LGA1156. BTW, I never heard someone who want to achieve 4.2GHz on LGA1156 but failed so far.(p.s. HT needs to be turned off at 4.2GHz for both LGA1156 and LGA1366 i7)Furthermore, not much voltage increase being required on LGA1366 for aggressive OC is due to its worse BJTs which has a lot of leakage current even when they are in the idle state. Since you have a lot of extra unnecessary current for LGA1366 (and therefore LGA1366 runs much hotter which is not good), you don't need to increase much voltage to raise the power for high OC according to P=VI.Read the article from the link below. It is about the new BJT & PCU(power control unit) on LGA1156 platform which acts as a TRUE switch. This means that there's almost no leakage current at idle state.http://www.anandtech.com/cpuchipsets...spx?i=3634&p=4[/citation]

Sir, to clarify further the X58 platform required more voltage to reach 4.00 GHz and consumed far more power in this comparison. In fact, the 870 test CPU operates at up to 4.3x GHz using air cooling with HT enabled while the 920 comes up short of 4.1 GHz.
 
[citation][nom]Crashman[/nom]Sir, to clarify further the X58 platform required more voltage to reach 4.00 GHz and consumed far more power in this comparison. In fact, the 870 test CPU operates at up to 4.3x GHz using air cooling with HT enabled while the 920 comes up short of 4.1 GHz.[/citation]
I was actually comparing 860 with 920, ignoring all the rest i7s due to their insane price.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom