Interested in torturing your 580 yet?

[Mistwalker]

Has anyone tested F@H performance on the 580 and confirmed whether throttling mechanisms impact it at all?

 

[BD231]

How is that? Someone achieved 1ghz core already.

AMD and nV are not limiting overclocking by doing this.

Oh ...

 

[notty22]

Read the article you linked. The card throttled due to the VRM's overheating. It's a cooling issue, not a determined attempt to change the way a independently developed application interacts with the card / system.

All they had to do was monitor current. Then it's protection. Detecting a particular piece of software in the driver? That's manipulation.

If you read the article it clearly states 2 things were going on there. AMD never even emailed them back.
They state the program furmark is on AMD list of powervirus.

http://www.geeks3d.com/20090925/ati...on-against-power-virus-like-furmark-and-occt/

The brand new Radeon HD 5870 also called Cypress has a hardware protection against overcurrent generated by FurMark or OCCT GPU.

That’s enough for AMD! For Cypress based cards, AMD has hard-wired the protection and has implemented a hardware solution to the VRM problem, by dedicating a very small portion of Cypress’s die to a monitoring chip. This chip monitors the VRM. If the chip detects a dangerous situation (overload), the chip will immediately throttle back the card by one PowerPlay level. Once the dangerous situation has disappeared, the card goes back to its performance level. In the case of FurMark or OCCT, this can continue to go back and forth as the VRMs permit.
With this hardware protection, the Radeon HD 5870 is fully protected against FurMark or OCCT. One consequence is that a software protection in the driver is not needed anymore and you can use your favourite stress test tool to torture your brand new Cypress-based graphics card.
This protection has another consequence: cards with cheap VRM will be more throttle back than cards with high quality VRM. This difference will impact the processing power of the card and should be visible in FurMark’s benchmark score.

 

[notty22]

Has anyone tested F@H performance on the 580 and confirmed whether throttling mechanisms impact it at all?

'

It does not. A few reviews compared Fold@home performance to the gtx 480.
http://www.legitreviews.com/article/1461/15/

For our test we ran the 264 Fs_coil work unit on all of the pairs of cards and found that the GeForce GTX 580 was 12.7% faster per step than a GeForce GTX 480. When we overclocked the shaders of the card up to 1740MHz we were able to improve performance and the seconds per step dropped by five seconds! With this nice overclock on the 512 shaders we were able to get a 22% F@H performance improvement over a GeForce GTX 480 reference design! As you can see, the GeForce GTX 580 is faster, quieter and uses less power than a GeForce GTX 480.

 

[Mistwalker]

'

It does not. A few reviews compared Fold@home performance to the gtx 480.
http://www.legitreviews.com/article/1461/15/

Awesome, thanks. I noticed a couple review sites (including AT) did have a Folding@Home benchmark, but didn't know if these comparisons held up without throttling over long periods, with a good variety of work units.

 

[notty22]

What would happen if you start furmark on a AMD card across three monitors ? Eyefinity , I'd imagine the power pull would increase, more pixels ? What about a eyefinity 6 card ?
Anyone want to test ?

 

[Phynaz]

http://www.geeks3d.com/20090925/ati...on-against-power-virus-like-furmark-and-occt/

The brand new Radeon HD 5870 also called Cypress has a hardware protection against overcurrent generated by FurMark or OCCT GPU.

You're still not reading your own links.

ATI cards protect themselves against over-current and over-temperature conditions as all electronic devices should.

ATI cards do not check that a particular application is running and modify their operational parameters based upon that check.

 

[OCGuy]

You're still not reading your own links.

ATI cards protect themselves against over-current and over-temperature conditions as all electronic devices should.

ATI cards do not check that a particular application is running and modify their operational parameters based upon that check.

Semantics. If the protection put in place by AMD still limits Furmark, it is the same thing. Both companies don't want that program run on their cards it seems.

 

[notty22]

You're still not reading your own links.

ATI cards protect themselves against over-current and over-temperature conditions as all electronic devices should.

ATI cards do not check that a particular application is running and modify their operational parameters based upon that check.

Its states right in the link at geeks they targeted those TWO APPLICATIONS.
Also.
Out of the links showing throttling approaches from AMD and Nvidia, only ATI cards throttled when they weren't supposed to.

 

[Idontcare]

Semantics. If the protection put in place by AMD still limits Furmark, it is the same thing. Both companies don't want that program run on their cards it seems.

Its also money.

Costs far more to develop, implement, & verify a hardware solution.

I think NV's solution is just a stop-gap to them getting a hardware solution implemented.

AMD and Intel did the same thing with CPU's. Intel did the hardware thing, AMD opted for BIOS but it wasn't so effective (remember those Athlon's burning up) until they got hardware implemented for over-temps.

Now Intel has over-current protection, AMD will too (soon).

Its an iterative business that is also cost-sensitive, NV took the cost-sensitive baby-step that AMD just skipped right pass.

The spirit of the effort is the same, and the end-goal is the same...stop having people destroy their otherwise perfectly good and functioning product in a manner that elevates warranty/RMA costs.

I love MSI Afterburner, and I wish there was a CPU version of it, but I also like the idea of hardware based current protection for the hardware. It lets me be stupid and save me from myself at the same time

 

[notty22]

I don't think either company is going to elaborate in depth how their throttling approach works. I saw the term algorithm used talking about ATI's implementation on Cypress. Thats one clue.
Also there is hardware implementation on Nvidia's solution, or they could retro this to the gtx 480. They can't.
http://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_580/3.html

three Texas Instruments INA219 sensors measure the inrush current and voltage on all 12 V lines (PCI-E slot, 6-pin, 8-pin) to calculate power.

 

[Phynaz]

Its states right in the link at geeks they targeted those TWO APPLICATIONS.
Also.
Out of the links showing throttling approaches from AMD and Nvidia, only ATI cards throttled when they weren't supposed to.

It says that? Here's what I see:

"ATI Cypress (Radeon HD 5870) Cards Have Hardware Protection Against Power Virus Like FurMark and OCCT"

In other words those are examples of applications that are protected against.

Also, please identify when a card is not supposed to throttle. Are some over the limit conditions acceptable while others are not?

Or we can just leave it as is, I don't think either of us is going to give an inch, and you know what they say about arguing on the Internet...

 

[notty22]

My opinion all along is this new intelligent current monitoring is a very good thing. Especially if it stops this 100%.
I notice right off, both evga E760, Silverstone psu's were common between 2 cases. A poorly manufactured cable ,at the pin/connector stage maybe ?
http://www.overclock.net/nvidia/770988-e760-classified-mobo-sli-gtx-480s.html

http://www.evga.com/forums/tm.aspx?&m=536713&mpage=1

 

[notty22]

Bunch of new slides out for Cayman.
Seems AMD has a new 'power containment system'

 

[RussianSensation]

Bunch of new slides out for Cayman.
Seems AMD has a new 'power containment system'

Sounds great. Looks like AMD and NV are really getting serious about enforcing a limit on the power draw in order to protect its customers. I wonder how many cards were prematurily RMA'ed as a result of Furmark.