• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Used a GTX 690 for 2 weeks...

Joined
Mar 31, 2014
Messages
1,533 (0.37/day)
Location
Grunn
System Name Indis the Fair (cursed edition)
Processor 11900k 5.1/4.9 undervolted.
Motherboard MSI Z590 Unify-X
Cooling Heatkiller VI Pro, VPP755 V.3, XSPC TX360 slim radiator, 3xA12x25, 4x Arctic P14 case fans
Memory G.Skill Ripjaws V 2x16GB 4000 16-19-19 (b-die@3600 14-14-14 1.45v)
Video Card(s) EVGA 2080 Super Hybrid (T30-120 fan)
Storage 970EVO 1TB, 660p 1TB, WD Blue 3D 1TB, Sandisk Ultra 3D 2TB
Display(s) BenQ XL2546K, Dell P2417H
Case FD Define 7
Audio Device(s) DT770 Pro, Topping A50, Focusrite Scarlett 2i2, Røde VXLR+, Modmic 5
Power Supply Seasonic 860w Platinum
Mouse Razer Viper Mini, Odin Infinity mousepad
Keyboard GMMK Fullsize v2 (Boba U4Ts)
Software Win10 x64/Win7 x64/Ubuntu
Just thought I would quickly share my thoughts...

Bought a second hand GTX 690 while I was on holiday a while back, cost me 100 quid. Interesting piece of hardware history to have and quite a nice looking card overall.

Replaced my GTX 670 FTW with the 690 the other week just to see how it would run, now I've replaced it with a 1070 Ti SC. Overall I was pretty impressed, the card is really fast in games where the SLI profile works well. Mostly played Rainbow 6 Siege and BF4, ran X-plane a few times as well.

X-plane doesn't use SLI but the decent clock speeds on the card made some difference relative to the 670, however I'm mostly frame buffer limited because of the very large texture sizes on orthophoto (satellite imagery) scenery.

BF4 for the most part felt fine, didn't play it that much, but I did notice the framerate would dip quite often and it didn't feel as smooth as I would expect for a given framerate.

In the built in R6 Siege benchmark the (average) framerate almost doubled for the same settings at 1080p from 86 to 160. Though they tended to fluctuate quite a bit and at times the microstuttering was really bad... 90 FPS would feel basically the same as 40 FPS. The 1070 Ti does much better in this regard.

My real problem with it though was the noise. I was surprised that the TPU benchmark put it so close to the 670 and 680... The 670 FTW uses a 680 PCB and cooler, and is surprisingly quite for a blower card, only if you run furmark with a boosted power limit does it actually become noticeably noisy. On the other hand the 690 was constantly above my CPU and case fans as soon as I ran any game. The primary GPU is closer to the PCI bracket so gets much worse airflow, under load the temperature difference between the two chips was usually 10-15 degrees. The 915 base clock is only just attainable with the stock fan profile, the primary chip gets well into the high 80s with normal loads.

Overall interesting card, but I never would have bought one for the $1000 launch price, even if I would water cool it I would much rather have had a pair of high end 680s.
 
Low quality post by Gungar
"Overall interesting card, but I never would have bought one for the $1000 launch price, even if I would water cool it I would much rather have had a pair of high end 680s."

Like any non stupid person.
 
Just thought I would quickly share my thoughts...

Bought a second hand GTX 690 while I was on holiday a while back, cost me 100 quid. Interesting piece of hardware history to have and quite a nice looking card overall.

Replaced my GTX 670 FTW with the 690 the other week just to see how it would run, now I've replaced it with a 1070 Ti SC. Overall I was pretty impressed, the card is really fast in games where the SLI profile works well. Mostly played Rainbow 6 Siege and BF4, ran X-plane a few times as well.

X-plane doesn't use SLI but the decent clock speeds on the card made some difference relative to the 670, however I'm mostly frame buffer limited because of the very large texture sizes on orthophoto (satellite imagery) scenery.

BF4 for the most part felt fine, didn't play it that much, but I did notice the framerate would dip quite often and it didn't feel as smooth as I would expect for a given framerate.

In the built in R6 Siege benchmark the (average) framerate almost doubled for the same settings at 1080p from 86 to 160. Though they tended to fluctuate quite a bit and at times the microstuttering was really bad... 90 FPS would feel basically the same as 40 FPS. The 1070 Ti does much better in this regard.

My real problem with it though was the noise. I was surprised that the TPU benchmark put it so close to the 670 and 680... The 670 FTW uses a 680 PCB and cooler, and is surprisingly quite for a blower card, only if you run furmark with a boosted power limit does it actually become noticeably noisy. On the other hand the 690 was constantly above my CPU and case fans as soon as I ran any game. The primary GPU is closer to the PCI bracket so gets much worse airflow, under load the temperature difference between the two chips was usually 10-15 degrees. The 915 base clock is only just attainable with the stock fan profile, the primary chip gets well into the high 80s with normal loads.

Overall interesting card, but I never would have bought one for the $1000 launch price, even if I would water cool it I would much rather have had a pair of high end 680s.

I've used SLI as well and the 690 is not really any different, but for many games its microstutter heaven. That's probably the lack of smoothness you perceived. Another source is the VRAM or rather lack of it and its bandwidth compared to the amount of core grunt you've got.
 
Yeah, at the end of the day SLI is a very crude system... I think the future with high bandwidth links like infinity fabric and NVlink will be much more useful for multi-GPU setups thanks to the possibility of sharing memory
 
Or just age... GTX 600 series are now 4 generations old. Sure it's a double stacked GTX 680, but it's still a very old graphic card. I could grab a R9-295X2 back in the day which was at the time the only card actually capable of doing 4K smoothly, but later decided not to and go with newer, slightly slower single card (GTX 980). Multi-cards have always been problematic and things haven't really changed much...
 
AFAIK both of these cards perform as they did new... Even if it's 4 generations it doesn't change the fact that a 670 handily beats a 1050 Ti... I think there are enough kepler cards floating around at very respectable prices for me to say they should be a consideration for anyone looking for a low-mid range card.

I can hardly complain about the performance of my 670. My main motivator for the upgrade was the fact that I was very memory restricted in X-plane and the fact that I am driving a 144Hz monitor for siege. If you are not after all ultra settings (a lot of gamers aren't) or higher than 1080p60 display rates, they are perfectly fine.

As for the 690 or any other old dual card setup, not recommendable, especially for the audience i mention above, framerate stability is almost always more important because it includes the audience of competitive gamers.

As far as multi card setups, they have always been problematic because SLI is a very crude system. Only the video output is exchanged between the cards, each card is rendering either individual frames or separate parts of frames. There is no exchange of pre-rasterised data so you are very restricted on how you can share the workload between two cards. That's why higher bandwidth links are so significant of an upgrade.
 
That's a great personal review of this iconic card. I'd like to have one in my NVIDIA collection too.

You might like to read my comparison review in my siggy. :)
 
AFAIK both of these cards perform as they did new... Even if it's 4 generations it doesn't change the fact that a 670 handily beats a 1050 Ti... I think there are enough kepler cards floating around at very respectable prices for me to say they should be a consideration for anyone looking for a low-mid range card.

I can hardly complain about the performance of my 670. My main motivator for the upgrade was the fact that I was very memory restricted in X-plane and the fact that I am driving a 144Hz monitor for siege. If you are not after all ultra settings (a lot of gamers aren't) or higher than 1080p60 display rates, they are perfectly fine.

As for the 690 or any other old dual card setup, not recommendable, especially for the audience i mention above, framerate stability is almost always more important because it includes the audience of competitive gamers.

As far as multi card setups, they have always been problematic because SLI is a very crude system. Only the video output is exchanged between the cards, each card is rendering either individual frames or separate parts of frames. There is no exchange of pre-rasterised data so you are very restricted on how you can share the workload between two cards. That's why higher bandwidth links are so significant of an upgrade.

It does ? I thought a 1050ti was around 780s performance?
 
It does ? I thought a 1050ti was around 780s performance?
A 680 is around a 1050 (2GB) and a 780 is a bit above a 1050 ti
 
It's very hit and miss because the 1050s have very weak memory subsystems compared to thr 680/670
 
There are a couple of videos on YouTube
 
Are there any 8 GB GTX 690? Because there were 4 GB GTX 670 and GTX 680...
 
Back
Top