Anyone buying a Titan?

Page 8 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

DooKey

Golden Member
Nov 9, 2005
1,811
458
136
We were reassessed for our property tax paid last year and there was an error made in our favour, we're getting reimbursed a respectable sum of money. My immediate though honestly was I could get a few of these with some of it and it wouldn't feel like I was out the money as it was unexpected money.

Managed to change my mind. These are going to get price cuts in my personal estimation, quickly enough that I'd feel like an idiot if I paid $1000 for them on day one.

Come on, Groove. Jump in the pool, the water is warm.
 

Grooveriding

Diamond Member
Dec 25, 2008
9,108
1,260
126
Come on, Groove. Jump in the pool, the water is warm.

:awe: Can't do it. Video cards are a want not a need but I can't reconcile them as a luxury worth $1000. Feels like going to the movies and paying $200 for a ticket.
 

Lepton87

Platinum Member
Jul 28, 2009
2,544
9
81
Looks like I'll have to wait until 20nm GPUs come along to make a real upgrade. :( After 2 years I can't even buy a faster single card for the same amount money that I paid for my cards, this is very sad. The only way to get better price/performance is going with Mulit-gpus again, but I've had enough multi-GPUs quirks. If I could find a good buyer for my cards I could probably settle for ASUS MATRIX 7970. I'd probably lose some and gain some, but overall it would be worth it. I really developed a particular dislike of Multi-GPUs solutions.
 

lambchops511

Senior member
Apr 12, 2005
659
0
0
After reading the review I was a bit disappointed with it's performance. I know it's the fastest single GPU but in the UK I could get 3 (yes 3!) 7970's GE's for just £60 (approx $70/$75) more. I don't have the money but if I did have unlimited funds I might buy 3 titan's for tri-SLI but otherwise not a chance in hell. 3x 7970's would annihilate one Titan as 2 completely demolish it.

I know this forum is dominated by games, but at this price point it is ridiculously cheap for researchers and academics. We would probably have to buy on the order of > 100 CPUs just to get a similar performance (depending on app). I suspect Titan would sell like hotcakes, but the majority of the sales would goto PhD students and researchers rather than gamers.
 

Piano Man

Diamond Member
Feb 5, 2000
3,370
0
76
I try to buy cards for around $250 that are significantly better than the one I currently own.

GTS 8800 -> 4850 ->5850 ->7950 (took AMD long enough to bring this price down). If and when Titan comes down to 250-300, I will have to see what it can do against my 7950 running at 1150MHz. Answer is probably no, but you never know.
 

NoStateofMind

Diamond Member
Oct 14, 2005
9,711
6
76
The Titan reminds me of this back in the day

thumb_GEF_7800_GTX_512MB_3qtr.jpg
 

Blastman

Golden Member
Oct 21, 1999
1,758
0
76
As for the validity, CF stuttering has been under investigation from many sites for years now, so this is nothing new. The 40+% gains on the other hand I have never seen before in case of the 7950. Maybe I missed it, that certainly is a possibility.

As for pricing: I just cancelled my two Titans that I ordered in a weak moment (580 SLI getting old...). What does that tell you?
This was just in response to those people who think CF is an alternative with such good value. It isn't. A single 7970@OC on the other hand very much is.
Has anyone tested stuttering on a 3 GPU Crossfire setup? According to this THG … test 3x GPU X-fire was much smoother than a 2x GPU setup (at least as far as frame rates go). Although that doesn't show stutter, it may be the case that a 3x GPU X-fire setup is much smother than 2x.
 

boxleitnerb

Platinum Member
Nov 1, 2011
2,601
2
81
Has anyone tested stuttering on a 3 GPU Crossfire setup? According to this THG … test 3x GPU X-fire was much smoother than a 2x GPU setup (at least as far as frame rates go). Although that doesn't show stutter, it may be the case that a 3x GPU X-fire setup is much smother than 2x.

THG tested with a CPU bottleneck, look at that poor scaling. And they only presented hard data for one single game. So it is at best inconclusive.

I think I will buy the Titans after all, because it occurred to me that I can still sell my 580s 3GB for 150-200 Euros a piece on Ebay. It is doubtful if I can do that at the end of the year should prices really have come down by then. So it really doesn't matter much if I pay 950-175 for a Titan now or if I pay 750-50 later. Maybe I can also get some 20 bucks for my waterblocks as well.
 

PowerK

Member
May 29, 2012
158
7
91
I am glad that I'm not the only one buying this overpriced piece of $hit. Heh.

My 690s are all sold and gone. Need new VGA.
I'll go for 2-Way or 3-Way SLI Titans depending on the availability & price here in Korea.
 

colonelciller

Senior member
Sep 29, 2012
915
0
0
has anyone seen benchmarks for high end software & CAD programs?

vue
avid
3ds max
maya
autocad
after effects
etc?
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
has anyone seen benchmarks for high end software & CAD programs?

vue
avid
3ds max
maya
autocad
after effects
etc?

just guessing a bit, but it should do really well in CUDA apps. It's using Geforce drivers though. So, it's not going to be the best choice for Max/Maya
 

aaksheytalwar

Diamond Member
Feb 17, 2012
3,389
0
76
I bought my 7970 in India at release for around $700-750 or so and I spent another $400+ on importing the after market cooler, paste etc with speedy delivery. And another $100 or so in mechanic fees, maybe more.

In India the Titan will probably release around $1300-1500 or so with no chance of the price coming down anytime soon. The 690 sells for $1200-1300 ATM here.

So definitely can't buy at the moment nor do I see much point in buying it either.

But a sub $700-800 amd card in 2014 may make sense should I not face any financial problems then, but for that it would need to be priced around $500-600 or less in USA.
 

Meghan54

Lifer
Oct 18, 2009
11,528
5,045
136
I must say, thanks Nvidia for the Titan release.

It really has returned this VC&G forum back to what made it more entertaining than P&N....with both the AMD apostles and Nvidia zealots about to have a holy war over video cards.

The chuckles have returned! Thanks!
 

Railgun

Golden Member
Mar 27, 2010
1,289
2
81
A workstation card that uses drivers optimized for those apps. (from either company)

For what...more than 3x the price?

No one is still getting what this card is (or isn't for that matter). But I digress...
 
Last edited:

lavaheadache

Diamond Member
Jan 28, 2005
6,893
14
81
I must say, thanks Nvidia for the Titan release.

It really has returned this VC&G forum back to what made it more entertaining than P&N....with both the AMD apostles and Nvidia zealots about to have a holy war over video cards.

The chuckles have returned! Thanks!

look who's up early! Nice to see ya
 

BoFox

Senior member
May 10, 2008
689
0
0
I went overboard on the sarcasm, I admit. I won't be doing that again. However, I definitely would like a feasible explanation of what's going on though - if SLI really is better than single GPU, hey great. I made the right choice.

Maybe toms and pcper just screwed the testing method up. I have no idea. Again: just a reasonable and logical explanation to put the chips in their place. What's going on?
With this chart from Toms:
bf3-2560-latency.png

We continue to see tiny gaps between frames from our single-GPU cards, though the GeForce GTX 690 consecutive frame time difference more than triples, on average. However, the latencies are still so small, and the frame rates so high, that we would still consider this a good result.
Well, judging from this chart (using the same data as the above bar chart) below:
bf3-2560-frot.png

Where GTX 690 is in fact averaging over 80fps, or less than 12.5ms.

The "average" (median) consecutive frame time difference is 14.1ms (from the first chart), which translates to 70.9 fps if taken by itself. (Actual average is a little over 80fps according to the blue line above.) So, with the ideal frame being only 12.5ms long (80fps), that means each frame would have to be alternating somewhere in between 25 (2 x 12.5) ms long and 0 ms long (or else it would be less than 80fps, period).

Somebody who graduated at the top of Algebra 2++, who isn't so rusty, please calculate the consecutive max and minimum frametimes for the two frames, if the consecutive frame time difference is 14.1ms, in order for the total to average 80fps! Show your work!

That's the ATF challenge (no, not AnandTechForum, but AlaskaThunder*uck)! :biggrin:
 
Last edited: