• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Hardware Canucks tests the Titan Z

Caporegime
Joined
9 Nov 2009
Posts
25,738
Location
Planet Earth
They managed to borrow a card from a developer for the review:

http://www.hardwarecanucks.com/foru.../66869-nvidia-titan-z-performance-review.html

An interesting result,which seems to pit the Titan Z slightly of the R9 295X2 consistently(no frametimes though).

Some of the results do seem to be a tad different from what PCPER saw:

http://www.pcper.com/reviews/Graphics-Cards/NVIDIA-GeForce-GTX-TITAN-Z-Review

I thought it was a tad odd,but due to platform differences. However it appears the driver stack used in the reviews was different:

PCPER:

Graphics Drivers AMD: 14.6 Beta
NVIDIA: 337.91 (GTX Titan Z)
NVIDIA: 337.88 (GTX 780 Ti)

HC:

AMD 14.4 Beta
NVIDIA 337.88 Beta

Still,I think the main issue as even HC says is price,and it appears to dump quite a bit of heat into the case,limiting its usage in smaller ones(although I suppose like with the R9 295X2 the cooler size is another issue here).

It will be interesting to see how a card based on a GTX780 or GTX780TI would perform.
 
Last edited:
I dont think anyone would argue its worst feature is the silly price its at.
But i must say it did better in the tests than i thought it would. While i wouldnt buy one i can see the apeal for some EG the owners of the Test card especially if they wanted a machine they could also game on.
 
Pretty cool, pity they couldn't over clock it though, shame Nvidia never gave any review samples out but can't really complain. I still wouldn't buy one though :P
 
They managed to borrow a card from a developer for the review:

http://www.hardwarecanucks.com/foru.../66869-nvidia-titan-z-performance-review.html

An interesting result,which seems to pit the Titan Z slightly of the R9 295X2 consistently(no frametimes though).

Some of the results do seem to be a tad different from what PCPER saw:

http://www.pcper.com/reviews/Graphics-Cards/NVIDIA-GeForce-GTX-TITAN-Z-Review

I thought it was a tad odd,but due to platform differences. However it appears the driver stack used in the reviews was different:

PCPER:

Graphics Drivers AMD: 14.6 Beta
NVIDIA: 337.91 (GTX Titan Z)
NVIDIA: 337.88 (GTX 780 Ti)

HC:

AMD 14.4 Beta
NVIDIA 337.88 Beta

Still,I think the main issue as even HC says is price,and it appears to dump quite a bit of heat into the case,limiting its usage in smaller ones(although I suppose like with the R9 295X2 the cooler size is another issue here).

It will be interesting to see how a card based on a GTX780 or GTX780TI would perform.

Why did they use old drivers for AMD? why did they not use 14.6 with the updated CF profiles?
 
It will be interesting to see how a card based on a GTX780 or GTX780TI would perform.


It is based on a 780ti basically with the DP not crippled so 2x Titan Blacks and that would not make any difference to the Temps as many of the Compute cards lower clocks to run DP mode (Titan did not sure on Titan Black).
 
Seeing the tests above, I wonder is 2 GTX780 3GB will cut through TESO without AA with the Acer 4K G-Sync monitor.

Given that they perform ok at 4K with 4 MSAA
 
So 2 780ti's can outperform the Titan z at 1440p which you can get for half the price ? Have they made this card just because they can ? :)
 
So 2 780ti's can outperform the Titan z at 1440p which you can get for half the price ? Have they made this card just because they can ? :)

They didn't strip out the DP performance on the Titan Black like they do with the 780TI, for that reason its worth the extra £800. or so some say, not my reasoning :)

Why did they strip out the DP performance in the first place? so they can charge you to put it back.

To be fair to Nvidia at least they are charging you more to put back what they removed to start with, and not the other way round. :D
 
Why did they use old drivers for AMD? why did they not use 14.6 with the updated CF profiles?

To be fair they are not using the latest Nvidia drivers either, though its a bit odd, you would think even if they had to rush the TZ back to it's owner and that meant they couldn't redo with newer drivers, the 14.6 betas have been out for over two weeks now so could have used them. Maybe their logic was if they didnt use betas for Nvidia they shouldn't for AMD either /shrug :confused:
 
To be fair they are not using the latest Nvidia drivers either, though its a bit odd, you would think even if they had to rush the TZ back to it's owner and that meant they couldn't redo with newer drivers, the 14.6 betas have been out for over two weeks now so could have used them. Maybe their logic was if they didnt use betas for Nvidia they shouldn't for AMD either /shrug :confused:

They should have used the latest drivers for both, its their job to show the true performance of the GPU's on the day, this isn't it.
 
Depends when they started their testing, they could have been 3/4 of the way through the amd stuff before the new drivers came out, honestly don't expect them to scrap all of that work and start from scratch again? They could be there forever doing it that way.

I think its important drivers are recent, not the absolute latest betas though, it would make testing and comparing a pain.

Always liked hardware canuks tbh, can't remember any of their GPU reviews but their case reviews are amongst the best :)
 
Sometimes newer drivers are worst or add support for 1 game and lower other games.

Best use last known good stable WHQL drivers to be fair, not 2 Betas.
 
Depends when they started their testing, they could have been 3/4 of the way through the amd stuff before the new drivers came out, honestly don't expect them to scrap all of that work and start from scratch again? They could be there forever doing it that way.

I think its important drivers are recent, not the absolute latest betas though, it would make testing and comparing a pain.

Always liked hardware canuks tbh, can't remember any of their GPU reviews but their case reviews are amongst the best :)

+1
 
Pretty cool, pity they couldn't over clock it though, shame Nvidia never gave any review samples out but can't really complain. I still wouldn't buy one though :P

That's a fair point, you can moan that Nvidia pulled a fast one by not handing out review samples to the press to avoid bad publicity but 99% of us were never going to buy one anyway.
 
Depends when they started their testing, they could have been 3/4 of the way through the amd stuff before the new drivers came out, honestly don't expect them to scrap all of that work and start from scratch again? They could be there forever doing it that way.

I think its important drivers are recent, not the absolute latest betas though, it would make testing and comparing a pain.

Always liked hardware canuks tbh, can't remember any of their GPU reviews but their case reviews are amongst the best :)

I agree about Canuks, yes. i have a lot of respect for them

I would always use the latest drivers, even if that means running the tests all over again every time, if i proclaim to inform my readers then i would be sure to give my readers the most up-to-date information available to me, otherwise its just poor quality journalism. Not something that i would be proud of, especially if i'm a professional being in one form or another paid for it.

IMO they let this one slip.
 
I agree about Canuks, yes. i have a lot of respect for them

I would always use the latest drivers, even if that means running the tests all over again every time, if i proclaim to inform my readers then i would be sure to give my readers the most up-to-date information available to me, otherwise its just poor quality journalism. Not something that i would be proud of, especially if i'm a professional being in one form or another paid for it.

IMO they let this one slip.

In an ideal world, yes. But the reality is often quite different. Especially if you've got deadlines.

The review isn't rubbish as a result it just has a caveat. The drivers which have been used are stated so I don't have any issue.
 
It's still not ideal, since due to the 3 slot design of the card in many boards you can't fit two in anyway but you'd be able to fit 2 Titans or 2 780s. Even in some boards where you could fit 2 you wouldn't be able to fit 4, which makes me question the advantage in compute tasks over a bank of Titans.

The only advantage is if you're sticking it into a board without SLI capability... which makes me wonder why the hell you'd spend 2 grand on a GPU to stick it into a 50 quid motherboard.

Really what everyone wants is a "GTX 790" in a 2 slot cooler that isn't 5x the price of a single card.
 
In an ideal world, yes. But the reality is often quite different. Especially if you've got deadlines.

The review isn't rubbish as a result it just has a caveat. The drivers which have been used are stated so I don't have any issue.

Why are you talking about the review being rubbish? thats a very strong word to use. do you think its rubbish? i disagree entirely, its not rubbish, its just not as good as it could be. :)
 
Why are you talking about the review being rubbish? thats a very strong word to use. do you think its rubbish? i disagree entirely, its not rubbish, its just not as good as it could be. :)

It was inferred in your post that it wasn't up to scratch. But sure get all hung up on one word. :D

Why are you asking if I think it's rubbish? My opinion on the matter is stated in the very same post which you've quoted. Generally it helps if you read.
 
Back
Top Bottom