[PCGH.de] Titan sli Vs 7970 Toxic cf - Page 21 - Overclock.net

Forum Jump: 
Reply
 
Thread Tools
post #201 of 666 Old 03-10-2013, 11:46 AM
Overclocker
 
Tonza's Avatar
 
Join Date: Jun 2010
Posts: 579
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 18

| Threadripper 1950X @ 4.1Ghz / Enermax Liqtech 360 TR4 | ASUS ROG Zenith Extreme | 32GB G.Skill Trident Z RGB 3200mhz CL14 | Geforce GTX 1080Ti FE @ 2100mhz / Alphacool Eiswolf GPX PRO /w 2x Corsair ML120| 1TB 850 PRO + 1TB 850 EVO | Xonar Essence STU + 2x Genelec G Two + Genelec 5050A | Phanteks Enthoo Evolv + EVGA Supernova G3 850W | Samsung S34E790C (34" 3440x1440 Curved Ultrawide)|
Tonza is offline  
Sponsored Links
Advertisement
 
post #202 of 666 Old 03-10-2013, 11:53 AM
New to Overclock.net
 
keikei's Avatar
 
Join Date: Feb 2013
Location: ketosis
Posts: 8,090
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 314
Quote:
Originally Posted by Tonza View Post


I'm sorry to laugh, but that was funny as hell! biggrin.gif I would imagine some people here have SLI Titans. Love to see some unbaised benchmarks. I dont care either way, just a real comparison.

keikei is offline  
post #203 of 666 Old 03-10-2013, 11:58 AM
PC Gamer
 
sherlock's Avatar
 
Join Date: Aug 2012
Location: Indiana, USA
Posts: 5,433
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 312
Quote:
Originally Posted by Cannon19932006 View Post

Regardless of the clocks, your looking at 2 $400 dollar cards well within obtainable overclock, almost matching up against 2 $1000 dollar cards at bare stock clock. It reiterates what we all know here on OCN.

1. Highest performance hardware usually costs a good premium (i7E a great example)
2. Overclocking can help lesser hardware obtain performance near the premium hardware.
3. AMD and Nvidia both have extremely impressive hardware for us this generation, hardware i compare to the release of the nvidia 8000 series.


We should all enjoy the advancements they have made as a whole, instead of fighting about it. Also get some of the free performance out of your card, whether it's amd or nvidia.


I would love to have either card lol, 7970 toxic at 1200, or Alatar's Titan at 1200+ devil.gif

Until the reason behind Titan's absurd 58% SLI scaling is found(20% worse than Techpowerup's review of the same game at same res), I wouldn't bother drawing any conclusion from this benchmark beside Nvidia needs to get some serious work done on Titan SLI driver support.
Quote:
Originally Posted by Lumo841 View Post

Why are people so admittedly defending the Titan...it got beat.

If the SLI/CFX(Clear win for Toxic CFX in term of FPS) results were consistent with the single vs single(Titan won) results that would be a non issue. But in this benchmark we have the worst Titan SLI scaling (barely 58%) in any review ever made with no explanation why.

I don't hug trees, but I do drive a Prius wink.gif

post-flame-small.gif GTX 1080 Overclock Leaderboard & Owners Club post-flame-small.gif


 
sherlock is offline  
Sponsored Links
Advertisement
 
post #204 of 666 Old 03-10-2013, 12:01 PM
slow TSMC = more telescopes
 
Alatar's Avatar
 
Join Date: Jan 2010
Location: Finland
Posts: 16,905
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 249
Quote:
Originally Posted by eXXon View Post

But there is when saying AMD has 'really' zero CF scaling and then linking a review that didn't officially publish their results yet and showed just one preview example …..

I agree this Thread/review needed to test the Titans @ stock vs normal GHz 7970s, but why would you try to advertise that AMD CF is nonexistent & push it as fact?

Took me a while to get it (I'm not very bright) but I'm guessing you didn't pay for the Titan in your sig.....or the 590 before it.....

I have never said any of what you are saying that I said. What I originally stated that 7950 CFX is great until you hit problems with the irrelevant frames.

I have also multiple times said that I do not believe that this is the case in all situations and I have not claimed that CF doesn't exist. However unless you want to refute PCPer's findings then their tests are accurate.

And as for your last comment, I'm just going to say that I work for my components. I also study, I would ask that you would stop making comments like that, for someone who actually doesn't have the money, the comment might be hurtful.
Quote:
Originally Posted by bencher View Post

Makes sense now thumb.gif
Newegg made an error then? Or everyone has your Titan.

Was the reviewer right about heat seeing as your are at 80c?

I already explained TPUs graphs about GPU boost 2.0 and I do not wish to do that again, please go see the discussion between myself and defoler. The crux of the argument is that the temperature of the card doesn't act like it does in TPUs graphs unless you lock the fan to a low RPM.

As for newegg listing them as 837MHz, yes, yes they do. Now how many times has it been proven by reviewers and users alike that that clock speed which is the absolute minimum base clock is not what the card runs at. Titan is designed to run in SFF systems, it will boost in extremely small enclosures and even with high temps. When it is in a normal case with good airflow, it will boost way beyond the 867MHz mark to around 960-993MHz, depending on the card. This isn't up for debate, that's how the card works.
Quote:
Originally Posted by sherlock View Post

If the SLI/CFX(Clear win for Toxic CFX) results were consistent with the single vs single(Titan won that) results that would be a non issue. But in this benchmark we have the worst Titan SLI scaling (barely 58%) in any review ever made with no explanation why.

And the CFX setup did not win, even with the absolutely massive frequency advantage and abnormally low SLI scaling.

Chimp Challenge 2 x Participant  15 Million+ Folding at Home points 

Alatar is offline  
post #205 of 666 Old 03-10-2013, 12:13 PM
I hate portals
 
eXXon's Avatar
 
Join Date: May 2012
Location: Mid East
Posts: 3,669
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 326
Quote:
Originally Posted by Alatar View Post

I have never said any of what you are saying that I said. What I originally stated that 7950 CFX is great until you hit problems with the irrelevant frames.

I have also multiple times said that I do not believe that this is the case in all situations and I have not claimed that CF doesn't exist. However unless you want to refute PCPer's findings then their tests are accurate.

And as for your last comment, I'm just going to say that I work for my components. I also study, I would ask that you would stop making comments like that, for someone who actually doesn't have the money, the comment might be hurtful.....

My bad for not understanding what you meant earlier.
smile.gif That makes more sense. Thank God you explained it or else I would not have slept tonight (seriously).

I can relax now headphone.gif

RAIL Does Not Mean POWER.

Look up this Database before buying a PSU.

post-flame-small.gif_.=5 GHz Overclock Club=._post-flame-small.gif
post-flame-small.gif_. I <3 Mechanical Keyboard._ post-flame-small.gif
eXXon is offline  
post #206 of 666 Old 03-10-2013, 12:20 PM
Programmer
 
bencher's Avatar
 
Join Date: Mar 2011
Posts: 4,122
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 81
Quote:
Originally Posted by Tonza View Post


lachen.gif

@ Altar, were you saying the same thing about clocks when the 680 boosting while the 7970 wasn't (Before ghz edition)?

I am will to bet my savings you weren't.
bencher is offline  
post #207 of 666 Old 03-10-2013, 12:21 PM
Intel Overclocker
 
ivanlabrie's Avatar
 
Join Date: Aug 2011
Location: Buenos Aires, Argentina. Land of the obscenely expensive hardware
Posts: 8,636
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 342
The level of biased claims in this thread and fanboyism is ridiculous...I'm gonna have to side with Alatar and I know it first hand (like any other Kepler user, or Titan) that stock boost on paper does NEVER reflect the boost you see in gpu-z or precision-x while using the card. It's normally a lot higher than that.

Also, I would have rather seen the guy test these on a test bench or even a cardboard box to hold the mobo in place with a fan or two moving air accross the board, and post REAL stock results and not ones where it seems he manipulated them on purpose cause of his bias or sheer stupidity.
I'm a bencher, and I go for whatever's faster single gpu wise, I could never afford a newer card currently but I'd get a single Titan if I could. For a gaming machine at 1080p I'd go with bang/buck of course (think single 7870 XT biggrin.gif)

I can't understand how you guys can't see the bias in this review, it just doesn't make sense at all. His testing methodology is definitely flawed. A proper review would have been:
-Run both cards at stock, compare results without further manipulation and using the latest drivers.
-Overclock both set of cards to the max stable clocks and post results (a few games, and a few synthetic benchmarks for good measure)
-Publish a graph comparing that to older results if possible, showcasing different 7970 variants, as well as 680s and whatnot.
ivanlabrie is offline  
post #208 of 666 Old 03-10-2013, 12:23 PM
slow TSMC = more telescopes
 
Alatar's Avatar
 
Join Date: Jan 2010
Location: Finland
Posts: 16,905
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 249
Quote:
Originally Posted by bencher View Post

lachen.gif

@ Altar, were you saying the same thing about clocks when the 680 boosting while the 7970 wasn't (Before ghz edition)?

I am will to bet my savings you weren't.

Never did claim that the 680 didn't boost higher than 1006MHz, or higher than 1058MHz for that matter. To do that would have been absurd.

Chimp Challenge 2 x Participant  15 Million+ Folding at Home points 

Alatar is offline  
post #209 of 666 Old 03-10-2013, 12:24 PM
I hate portals
 
eXXon's Avatar
 
Join Date: May 2012
Location: Mid East
Posts: 3,669
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 326
Quote:
Originally Posted by bencher View Post

lachen.gif

@ Altar, were you saying the same thing about clocks when the 680 boosting while the 7970 wasn't (Before ghz edition)?

I am will to bet my savings you weren't.

They didn't leave the Titans @ stock, that's the point. They should have let it run without doing anything to it, hence stock.

No one did a review for the 680 when it came out without the boost function. But here, they removed it for no reason.

RAIL Does Not Mean POWER.

Look up this Database before buying a PSU.

post-flame-small.gif_.=5 GHz Overclock Club=._post-flame-small.gif
post-flame-small.gif_. I <3 Mechanical Keyboard._ post-flame-small.gif
eXXon is offline  
post #210 of 666 Old 03-10-2013, 12:27 PM
Networking Nut
 
TwilightEscape's Avatar
 
Join Date: Apr 2007
Location: Hot as hell, New Mexico
Posts: 826
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 72
Quote:
Originally Posted by bencher View Post

lachen.gif

@ Altar, were you saying the same thing about clocks when the 680 boosting while the 7970 wasn't (Before ghz edition)?

I am will to bet my savings you weren't.

Because the ghz edition is an overclocked 7970?

How difficult is it for people to get what comparing of stock cards is? Think reference design. No overclocking, no overvolting, no undervolting, just what the reference cards are out of the box. Why would someone complain about benches between the 7870 at stock against the 680 at stock? Makes no sense. "Oh let me overclock this card by 30% with better cooling and and leave the other at reference design." This seems logical.



Run benches of a reference Titan and a reference 7970, THEN you can add overclocked results as well. Otherwise it is meaningless to overclock one card a ton and actually reduce the clocks on the other.

TwilightEscape is offline  
Reply

Quick Reply

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


Forum Jump: 

Posting Rules  
You may post new threads
You may 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