GTX 780 Ti reviews

[skipsneeky2]

I think this Titan blower design from nvidia couldn't really get any better, as far as blowers goes anyways. Personally i had come to learn that reference coolers are bad news for high end gpus from personal experience but i have to say they did it right this time with this design.

The design on the 670....wow don't get me started on that pos as i took one apart.

 

[railven]

I think this Titan blower design from nvidia couldn't really get any better, as far as blowers goes anyways. Personally i had come to learn that reference coolers are bad news for high end gpus from personal experience but i have to say they did it right this time with this design.

The design on the 670....wow don't get me started on that pos as i took one apart.

I remember reading issues with the 670's, did that issue transfer over to the 660 Ti? I recently RMA'd a ref 660 Ti with the small PCB (I only used it for 3 days before it bit the ghost, but I got it used (for a great price) so not sure if previous user or shipping may have accelerated it's death.)

Will be picking up a second one today for SLI but a little worried. I've never had a card fail while it was in my ownership. Then I saw some 660 Ti's came with a full PCB not the half one that the ref uses.

 

[AdamK47]

I did a quick google search and found this review describing the open end of the reference cooler.

http://www.overclockersclub.com/reviews/nvidia_gtx780ti/

Found other reviews of the 780 and Titan that describe it as an intake. It makes sense given the tight spaces of SLI. I've stressed my cards and have only felt the heat exiting out of the back. In fact, I use the GPU shroud from my HAF-X and point a Corsair SP120 directly into the open end of my Titans. It helps keep the cards cool even when stress testing. I've seen a 5 degree difference on the top card with the shroud/fan on.

 

[BallaTheFeared]

 

[skipsneeky2]

I remember reading issues with the 670's, did that issue transfer over to the 660 Ti? I recently RMA'd a ref 660 Ti with the small PCB (I only used it for 3 days before it bit the ghost, but I got it used (for a great price) so not sure if previous user or shipping may have accelerated it's death.)

Only issue i had with the 670 reference was this......

This isn't worthy of a argument for being a crap ass heatsink and fan, the 290 and the Titan coolers are 4x as better and hell the whole argument concerning the Titan cooler is completely laughable when there is years worst out there.

 

[SlowSpyder]

Based on the way I understand blowers to work, I have to think air would also be pushed out the back. With that being said, it is probably a lot smarter design by Nvidia. Based on the size of the rear heat sink I think it is safe to say we're not dealing with a lot of really hot circuitry on the back of the card. But even if it only increases the air temperature a few degrees, why push that over the GPU heat sink at all if you don't have to? It would likely be an inconsequential amount of heat dumped in the case, but might be enough to keep your GPU temps just a few degrees lower, which of course means better boost clocks and better overclocks.

All in all, nice job by Nvidia on the design.

 

[Gloomy]

Based on the way I understand blowers to work, I have to think air would also be pushed out the back. With that being said, it is probably a lot smarter design by Nvidia. Based on the size of the rear heat sink I think it is safe to say we're not dealing with a lot of really hot circuitry on the back of the card. But even if it only increases the air temperature a few degrees, why push that over the GPU heat sink at all if you don't have to? It would likely be an inconsequential amount of heat dumped in the case, but might be enough to keep your GPU temps just a few degrees lower, which of course means better boost clocks and better overclocks.

All in all, nice job by Nvidia on the design.

You also have to take into account situations where the front of the card is blocked, like in SLI configurations. In that case I doubt any air would be blown out into the case, but rather sucked in from the end of the card and exhausted out maybe (?)

 

[PurolatorHelp]

Purolator is holding mine hostage.. it went from LA to Vancouver over the course of 4 days. Based on where I live, they shipped it about 10% further away from me.

Hey Spidre,

Sorry about that. Shipments being shipped from LA usually transit in bulk with other packages generally for 2 business days and are then inducted into Canada at our Vancouver, BC depot. From there, packages are separated into individual shipments and sent off according to the destination and service selected. If you’d like me to verify when you can expect the new card, you can email the tracking # to customer.care@purolator.com.

Thanks,

Jérémie
Social Media Coordinator / Customer Care Team

Purolator Inc.
E-mail: customer.care@purolator.com
www.Purolator.com

Follow us on Twitter:
http://twitter.com/purolatorhelp

 

[RaulF]

Hey Spidre,

Sorry about that. Shipments being shipped from LA usually transit in bulk with other packages generally for 2 business days and are then inducted into Canada at our Vancouver, BC depot. From there, packages are separated into individual shipments and sent off according to the destination and service selected. If you’d like me to verify when you can expect the new card, you can email the tracking # to customer.care@purolator.com.

Thanks,

Jérémie
Social Media Coordinator / Customer Care Team

Purolator Inc.
E-mail: customer.care@purolator.com
www.Purolator.com

Follow us on Twitter:
http://twitter.com/purolatorhelp

Oh snap.

 

[RaulF]

I did a quick google search and found this review describing the open end of the reference cooler.

http://www.overclockersclub.com/reviews/nvidia_gtx780ti/

Found other reviews of the 780 and Titan that describe it as an intake. It makes sense given the tight spaces of SLI. I've stressed my cards and have only felt the heat exiting out of the back. In fact, I use the GPU shroud from my HAF-X and point a Corsair SP120 directly into the open end of my Titans. It helps keep the cards cool even when stress testing. I've seen a 5 degree difference on the top card with the shroud/fan on.

:wub:

What cooler is that for the CPU? Are you using push or pull method?

I have the same case. Feel free to pm me.

Sorry for off topic admins.

 

[Cloudfire777]

Finally, to remove the collected and dissipated heat, a rear-mounted fan constructed from an acoustic-dampening material blows it out through the exhaust and away from the card’s components. In the GTX 690, this fan was mounted centrally, pushing heat out into the chassis, increasing overall system temperatures. This was necessitated by the front and rear placement of the GPUs, and the need to remove the considerable amount of heat as quickly as possible. With only one forward-mounted GPU, lower power consumption, and the aforementioned improvements, the GTX TITAN was able to utilize a traditional rear-mounted fan, reducing temperatures and enabling faster clock speeds.

Source: Nvidia.com

 

[Silverforce11]

It would have to blow air both ways, but the air coming to into the case would probably be close to ambient, since the vast bulk of the heat is coming from the GPU and ram itself. As such it would not affect your system temps in any meaningful way.

It is a very good design in that its quiet for the cooling achieved. But only at stock, because all blowers get noisy once you OC. http://www.tomshardware.com/reviews/radeon-r9-290-review-benchmark,3659-18.html
Look at that vid, and tell me the 780 at 80% is not noisy. Thats just keeping the stock card from thermal throttling its top tier boost clocks..

Its their limitation unless they go massive size like the HIS blowers.

ps. Did i understand it how Tom gets the data for the review? The 780 is run at a manual 80% fan speed to ensure it does not throttle and maintain a consistent boost clock..

 

[Spidre]

Hey Spidre,

Sorry about that. Shipments being shipped from LA usually transit in bulk with other packages generally for 2 business days and are then inducted into Canada at our Vancouver, BC depot. From there, packages are separated into individual shipments and sent off according to the destination and service selected. If you’d like me to verify when you can expect the new card, you can email the tracking # to customer.care@purolator.com.

Thanks,

Jérémie
Social Media Coordinator / Customer Care Team

Purolator Inc.
E-mail: customer.care@purolator.com
www.Purolator.com

Follow us on Twitter:
http://twitter.com/purolatorhelp

Hey thanks!

 

[SlowSpyder]

You also have to take into account situations where the front of the card is blocked, like in SLI configurations. In that case I doubt any air would be blown out into the case, but rather sucked in from the end of the card and exhausted out maybe (?)

I don't believe that is how blowers work. They are more or less 'slinging' the air out the blades. The air comes in from the opening on the top. The air cannot go in all directions because the top and bottom are closed. But the front where the connections are and the GPU heat sink is located will obviously be open to allow the air to flow, and in this case the back of the card as well. By doing it this way, even if Nvidia is dumping a percentage of heat back in the case, we can judge from the size of the heat sink on the back of the card that it is likely very little. But it would probably be enough that the GPU is allowed to reach higher boost speeds and higher overclocks by at least smallish amount.

Regarding SLI/CF, I don't really think the cards heat up more because the space between the cards isn't big enough to pull air, at least I don't think that's the main contributing factor. We're not dealing with huge CFM's of air here. My bet would be the incoming air is warmed up by the warm PCB right on top of the bottom card as well as the warmth from the second card that is to be cooled. And possibly a smallish decrease in the amount of air that is easily pulled in due to the limited space. But my guess is that it is a combination of factors.

It is a smart design for a card that is meant to boost based on thermal and/or power use characteristics.

 

[3DVagabond]

It would have to blow air both ways, but the air coming to into the case would probably be close to ambient, since the vast bulk of the heat is coming from the GPU and ram itself. As such it would not affect your system temps in any meaningful way.

It is a very good design in that its quiet for the cooling achieved. But only at stock, because all blowers get noisy once you OC. http://www.tomshardware.com/reviews/radeon-r9-290-review-benchmark,3659-18.html
Look at that vid, and tell me the 780 at 80% is not noisy. Thats just keeping the stock card from thermal throttling its top tier boost clocks..

Its their limitation unless they go massive size like the HIS blowers.

ps. Did i understand it how Tom gets the data for the review? The 780 is run at a manual 80% fan speed to ensure it does not throttle and maintain a consistent boost clock..

Well, it sure seems that way from what they printed.

Noise Comparison with the Nvidia GeForce GTX 780

Same gaming loop, different graphics card. The GeForce GTX 780 at stock speed and settings is a lot quieter, but it pays for it by reaching its thermal limit quickly. The fan needs to be pushed quite a bit to achieve consistent GPU Boost frequencies. Seventy percent are enough for a cold card, but once it’s warmed up, a fan speed of 80 percent is needed to maintain those higher clock rates. This is the only way to get an apples-to-apples comparison of the two competing graphics cards.

GeForce GTX 780 - Metro Last Light Gaming Loop - 52 Percent Fan Speed (Thermal Throttling)

GeForce GTX 780 - Metro Last Light Gaming Loop - 80 Percent Fan Speed (No Thermal Throttling)

:\ WTH?

Makes their big expose' about the retail 290's throttling seem really strange.

 

[Abwx]

At least we know what "apples to apples" means for THG....

 

[3DVagabond]

At least we know what "apples to apples" means for THG....

I don't know what's worse, boosting nVidia's fan speed to stop throttling, or cutting AMD's to exaggerate it. Whatever it takes I guess to get the desired result. I hope we are misinterpreting something here, because it's starting to look like a complete whitewash from these sites in question.

We have AT's own numbers showing the two cards being basically equal when the 290X is run in uber mode. We have Hardware.fr's numbers showing exactly the same thing when they run the 290X uber and don't increase the cooling on the 780ti. Now we have Tom's talking about running the 780 @ 80% fan speed to prevent throttling after they hammer AMD and include benchmarks from a badly throttling card, bringing into question AMD's integrity, causing people to run around accusing AMD of cheating (again).

 

[Keysplayr]

Were they talking about 780 or 780Ti?
And either way, why would it need to maintain its boost clock? All it needs to do is maintain its base clock. And when it does throttle its only at max a couple of dozen MHz anyway. Not seeing why they opted to up the fan speed to maintain boost clock. Just run the card as intended out of the box.

 

[Silverforce11]

Were they talking about 780 or 780Ti?

Both: http://www.tomshardware.com/reviews/geforce-gtx-780-ti-review-benchmarks,3663-18.html

 

[Abwx]

Both: http://www.tomshardware.com/reviews/geforce-gtx-780-ti-review-benchmarks,3663-18.html

Curiously variability with retail and typical conditions
are no more an issue...

 

[Silverforce11]

Were they talking about 780 or 780Ti?
And either way, why would it need to maintain its boost clock? All it needs to do is maintain its base clock. And when it does throttle its only at max a couple of dozen MHz anyway. Not seeing why they opted to up the fan speed to maintain boost clock. Just run the card as intended out of the box.

Its so they can reach its optimal performance, since people who folk out that much $$ for enthusiast GPUs are likely to try and squeeze out the max performance they can get..

Either that, or an insidious plot whereby they artifically bench NV cards showing their best while again artificially benching AMD cards to show their worst..

Which you believe, is up to you.

 

[Keysplayr]

Which do you believe?

 

[Silverforce11]

Which do you believe?

I'd like to believe all the reputable review sites that have maintained a long and prosperous presence on the net, are neutral and always aim to put the best interest of their readers ahead of anything else.

 

[blackened23]

Kepler GPUs tend to boost far higher than advertised out of the box, and if they do throttle - in my experience they will throttle only 1-2 bins even in incredibly hot ambient temps. Meanwhile, 290X is throttling down to 727mhz in "quiet" mode. On the note of boosts, my 780 advertises a 1020 boost, yet out of the box it boosts to nearly 1100. So if I throttle by 1-2 bins (26mhz) what does that mean? I'm still ahead of the advertised boost.

I have never had issues with throttling with any Kepler GPU causing a performance drop even with auto fan. I suppose some want to obfuscate the issue, but it isn't an issue. Kepler variance is 1-2% at best. Meanwhile you have the 290X with near 20% variance between press/retail boards and throttling by 100s of MHz. These two situations aren't remotely similar despite any attempts to obfuscate it.

 

[Silverforce11]

If Kepler variance is 1-2% at best, maybe someone should tell TOMS that.. because 1-2% to me sounds REALLY SMALL and within the margins of errors on benches.