New GPU (Radeon 6570) - Need Help Interpreting Stress Test Results - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 9 Old 08-31-2012, 04:39 PM - Thread Starter
Member
 
snowboarder33's Avatar
 
Join Date: Dec 2011
Posts: 23
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hello all,

Recently I've decided to build a HTPC and to go the route of an Intel G540 + discrete GPU. Usage being: running XBMC from NAS, streaming mostly 1080p content. I have been debating between the Geforce GT 430, Radeon HD 6450, and the Radeon HD 6570 - one caveat, the card needs to be low-profile. All can be had in a similar price range, maybe the 6450 a little cheaper, but since this is going into a HTPC, I need the card to be fairly silent. Passive-cooled cards have interested me, only be cause I've read several reviews of people complaining about fan noise, however, on the other hand with passive card, people complain about it running very hot. I had the chance to purchase this card, HIS H657HS2G Radeon HD 6570 2GB 128-bit DDR3, for $40 after rebate, along with the option to return it within 60 days, so I thought why not test it out and see how it goes. Looked at some professional online reviews and saw the 6570 usually beating out the 430, so with the card being passive and low-profile, I thought it would be perfect.

Well I got the system up and running - just sitting out on a desk - and I ran some stress tests on the card, but I really have no idea how to interpret them or what to take from them. I do know, however, that the FPS I've been averaging is horrendous.

First, here are the results of a 3DMark Performance test:
FWtCO.jpg

Results of 3DMark Extreme test:
vqciO.jpg

I do understand comparability, so on 3DMark's website, they have a history of results users have submitted with near identical hardware, which can be seen here:
r6MBa.jpg

So it looks like my results are actually below the lowest. I did notice my card is only clocked at 1000Mhz, which is below the average of ~1334MHz, so I'm not sure how much of an effect that has.

Next, I did a Furmark 1920x1080 stress test for 15 min.
uBwii.jpg

Still the card was just sitting on my desk with no fans. And actually an update to the Furmark score, I have been running the thing for the past 25 minutes and it seems to top out at 76 C.


So I guess I have a few concerns. Given these extremely poor results, should I be worried of the performance of the card? Are the temperatures within reason or are they a cause for concern? Any other tests I should run?


Oh and I built this thing today (first time ever building one) so it's a fresh install of Win7 and downloaded the latest AMD driver. Would really appreciate any input/suggestions/feedback/pointers/anything - feeling a bit deflated by these results.

Thanks!
snowboarder33 is offline  
Sponsored Links
Advertisement
 
post #2 of 9 Old 09-01-2012, 05:17 AM
Member
 
Mnoe's Avatar
 
Join Date: Aug 2002
Location: VA, USA
Posts: 21
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I have that exact card in a test HTPC machine paired with some "old" AMD gear, connected to Onkyo NR906, on to an LG HDTV. I use jRiver MC/LAV/madVR. This card is fantastic as an HTPC, sips power, runs cool, handles everything I throw at it except 2160p DXVA (fine with software decode). I have no idea about those numbers you're showing, did you buy the card to run benchmarks? wink.gif
Mnoe is offline  
post #3 of 9 Old 09-01-2012, 06:51 AM - Thread Starter
Member
 
snowboarder33's Avatar
 
Join Date: Dec 2011
Posts: 23
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Good to know! That's reassuring.

Yes. Strictly for benchmarks - just kidding. I was just running them to see how the card performs. Do you possibly have some temperature readings of your card, maybe during 1080p playback, idle, etc.?
snowboarder33 is offline  
post #4 of 9 Old 09-01-2012, 07:31 AM
Member
 
Mnoe's Avatar
 
Join Date: Aug 2002
Location: VA, USA
Posts: 21
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I have it in an Antec Fusion case (oldler gen I), two fans, one for case exit and another for the CPU cooler (mini-Ninja), both spinning around 800RPM. CPU is a severely undervolted/underclocked Athlon II (idles at 700Mhz @ .75V, thx to K10Stat). Speedfan shows GPU idle temps anywhere from 45-51C. Playback temps never get above 62C, any format. Normally, I don't use DXVA for anything except VC-1 and that's using copy-back mode from LAV/madVR.

fwiw, machine also has an Intel GbE card, SSD and 1TB HD.
Mnoe is offline  
post #5 of 9 Old 09-01-2012, 08:13 AM - Thread Starter
Member
 
snowboarder33's Avatar
 
Join Date: Dec 2011
Posts: 23
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Great to hear, that's exactly what I wanted to know. That's about what my card is averaging as well.

Given your results with the card, I think it might be a keeper! Thanks!
snowboarder33 is offline  
post #6 of 9 Old 09-01-2012, 08:18 AM
Senior Member
 
dksc318's Avatar
 
Join Date: Apr 2012
Location: San Jose,CA
Posts: 453
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
76deg C is a bit high. 60deg C is highest you want to touch. Many newer intel CPU are spec'ed to 70deg C. case temperature.

I know you said you've selected a passive cooling card for low noise. A PWM or thermal controlled fan should be pretty quiet when things are cool. 40mm or 60mm fans are available with PWM. Look for them on the CWC parts website. Here is a small thermal controller mounted to a PCIe card.

The green tip touching the heatsink is a thermistor. This controller came out of an old ATX PSU and can control 3 fans.
dksc318 is offline  
post #7 of 9 Old 09-01-2012, 11:43 AM - Thread Starter
Member
 
snowboarder33's Avatar
 
Join Date: Dec 2011
Posts: 23
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Well are you taking into consideration it was a Furmark test at 1920x1080 for 15 minutes with no fan or are you saying that is high for that test? Because that was how it got up to the 76C mark, and as I mentioned, even remained at that temp up to 30 minutes. And isn't the Furmark stress test a lot more graphics intensive than watching a bluray, which is what I actually plan to use it for? Or does the Furmark test equate to a 1080p movie? Not really sure.

But last night I played a 20GB mkv movie file and it averaged around 44-46C.
snowboarder33 is offline  
post #8 of 9 Old 09-01-2012, 02:05 PM
AVS Special Member
 
GreenEyez's Avatar
 
Join Date: Jan 2010
Posts: 2,258
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 20
76 Celsius is fine, GPUs are different from CPUs (where the maximum temp is indeed around 70 degrees Celsisus) and can go up to 90 degrees Celsius when it comes to the maximum temps allowed. And yes, Furmak is more intensive than any modern 3D game, let alone 1080p which is nothing compared to Furmak. Running Furmak and Prime 95 (to stress the CPU and memory) at the same time is one of the most demanding tests you can put your machine through, if it handles that, it can handle anything .
GreenEyez is offline  
post #9 of 9 Old 09-01-2012, 07:29 PM - Thread Starter
Member
 
snowboarder33's Avatar
 
Join Date: Dec 2011
Posts: 23
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Wow, alright so Furmark is pretty intense - that's kind of what I was thinking. Then, so far, it looks like this card has fairly adequate cooling, at least in an HTPC sense. I have just been a bit apprehensive with the card since it was given some less than stellar reviews on Newegg and a few other websites.

Once I get everything installed in the case though, then I can run a few more tests and get a better feel as to what the actual temperatures will be, and now I have some knowledge of what temperatures to look for. And I'll give the Furmark+Prime95 test a shot, just to see if it highlights any major problems with the setup/hardware.

Thanks!
snowboarder33 is offline  
Reply Home Theater Computers

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off