Thursday, March 14th 2013

NVIDIA Working on Faster GeForce GTX 650 Ti

In an attempt to avert the threat posed by the upcoming AMD Radeon HD 7790 "Bonaire," NVIDIA is reportedly changing the specifications of its current GeForce GTX 650 Ti. The company could increase reference clock speeds of its current GTX 650 Ti, without tinkering with physical specifications such as CUDA core count (of 768) or memory bus width of 128-bit. The GTX 650 Ti currently features reference clock speeds of 925 MHz core, and 1350 MHz (5.40 GHz GDDR5-effective) memory. The rehashed GTX 650 Ti could be released towards the end of March, or early-April.
Source: SweClockers
Add your own comment

44 Comments on NVIDIA Working on Faster GeForce GTX 650 Ti

#1
tastegw
Gpu wars are good, they need more battle on prices right now on all levels.
Posted on Reply
#2
las
Why? There are alot of 650 Ti OC cards out. I don't get it :)
Posted on Reply
#3
EzioAs
Actually, I think they should increase the memory bus. The GPU is actually pretty strong considering it has a 128-bit bus.
Posted on Reply
#4
btarunr
Editor & Senior Moderator
lasWhy? There are alot of 650 Ti OC cards out. I don't get it :)
Because when HD 7790 launches, people won't compare it to those OC cards, but the reference design. Even then, I think it's too late. It's improbable that TPU will compare HD 7790 to the new 650 Ti.
Posted on Reply
#5
remixedcat
lasWhy? There are alot of 650 Ti OC cards out. I don't get it :)
I got the EVGA SSC one
Posted on Reply
#6
Abate
Re:
lasWhy? There are alot of 650 Ti OC cards out. I don't get it :)
Did you ask the same question, when AMD released 7970GE?
Posted on Reply
#7
racedaemon
Great, juts when i was about to buy one. At least sell the 2GB one for cheaper.

I noticed the 650Ti was on sale on several online shops but i believed there is no way a replacement would come, so remember, when two or more retailers put on sale a product it means only one thing.
Posted on Reply
#8
xvi
More granular binning means the original 650 Ti, which probably won't change in price, will be less likely to overclock, wouldn't it?
Posted on Reply
#9
RCoon
racedaemonso remember, when two or more retailers put on sale a product it means only one thing.
They're competing against each other? Business in that market segment is slow? They got a deal on stock from somewhere?

Can mean a lot of things, certainly not when two shops have a gpu on sale, a new one is coming out in a month.
Posted on Reply
#10
Casecutter
xviMore granular binning means the original 650 Ti, which probably won't change in price, will be less likely to overclock, wouldn't it?
Exactly! Just means more AIB Über renditions. We already know that cards like the Zotac are running memory clocks like 1550Mhz (6200Mhz effective) pushing the bandwidth to 99.2Gb/s gives a decent kick in the pants you just had to open the wallet to get it.

Here's the problem, Über versions today don't get close to a 7850 1Gb, on Catalyst 13.1 the spread is like 24%. So if 7790 are 10% below... that tells me even this gerrymandering will still give AMD a 15% lead. But it furnishes Nvidia opportunity to, after all the 7790 reviews hit, say "but wait we got faster" ... but then it's to late 7790 review are done, while not sending-out any to reviewers to vie in the ring. Playing the pretense "we've answered the competition".
Posted on Reply
#11
rainzor
I thought they were going to release GTX660SE (768cores, 192bit/24rop part), not this bs :/
Posted on Reply
#12
Kaynar
AbateDid you ask the same question, when AMD released 7970GE?
Marketing trick to show that the 7970 is faster than GTX680. Cause it seems most people can't use the mouse to slide 2 bars from 925/1375 to 1125/1500.
Posted on Reply
#15
Prima.Vera
I think they should release the new generation already. Time to say "goodbuy" to that 58xx crossfire... ;)
Posted on Reply
#16
btarunr
Editor & Senior Moderator
rainzorI thought they were going to release GTX660SE (768cores, 192bit/24rop part), not this bs :/
They won't do such a thing because:-
  • It's the $140~$150 price-point that HD 7790 is after, and which GTX 650 Ti holds
  • "GTX 660 SE" would step up production cost (192-bit == at least six memory chips)
  • Tinkering with clocks on the existing 650 Ti costs nothing.
  • Something else that wouldn't cost anything, is giving GTX 650 Ti the GPU Boost feature, something it currently lacks
Posted on Reply
#17
el fore
Reply

The war of GPUs continuous eh.

Many articles and post I saw saying that PC gaming is dead.

The PC industry will never end, NEVER.
Posted on Reply
#18
Delta6326
btarunrThey won't do such a thing because:-
  • It's the $140~$150 price-point that HD 7790 is after, and which GTX 650 Ti holds
  • "GTX 660 SE" would step up production cost (192-bit == at least six memory chips)
  • Tinkering with clocks on the existing 650 Ti costs nothing.
  • Something else that wouldn't cost anything, is giving GTX 650 Ti the GPU Boost feature, something it currently lacks
I would like to see boost, and 2GB, I think it would really make for a sweet spot.
Posted on Reply
#19
racedaemon
btarunrIt's the $140~$150 price-point that HD 7790 is after, and which GTX 650 Ti holds
Just curious. I see you are from India, is the 650Ti ~$140 there, as in the US? In Romania it is ~$200. Do only people in Europe get screwed on hardware prices?
Posted on Reply
#20
EzioAs
racedaemonJust curious. I see you are from India, is the 650Ti ~$140 there, as in the US? In Romania it is ~$200. Do only people in Europe get screwed on hardware prices?
I can guarantee it's not only Europe, but Asia as well
Posted on Reply
#21
tokyoduong
racedaemonJust curious. I see you are from India, is the 650Ti ~$140 there, as in the US? In Romania it is ~$200. Do only people in Europe get screwed on hardware prices?
EzioAsI can guarantee it's not only Europe, but Asia as well
It's everywhere in the world except the US. Don't complain when it's your country's tax system at work causing cost push price inflation.
Posted on Reply
#22
HumanSmoke
btarunrEven then, I think it's too late. It's improbable that TPU will compare HD 7790 to the new 650 Ti.
Giving the card a boost capability and adding a stupid naming convention (e.g. GTX650Ti Boost) would seem enough to get a card retested and added to a benchmark comparison.

Alternatively, higher clocks, and renaming it the GTX655Ti would also likely qualify.

Would seem extremely unlikely for Nvidia to drop the ball on a marketing opportunity.
Posted on Reply
#23
Casecutter
btarunrSomething else that wouldn't cost anything, is giving GTX 650 Ti the GPU Boost feature
Well, I don't know if it will cost nothing?

Nvidia or AIB's have to design a PCB because isn't there a few PCB components like those INA219 power monitor chips to regulate Boost to pack on that smurf board? Sure they could move to the large GTX660, losing the one nice feature the diminutive size... would they add SLI. While I wonder if they did go with Boost what's the power section going to need? There's production build-up for PCB's by the AIB's, new printing of packaging, marketing… costs?

I think Nvidia just let loose more "higher" binned chips for less money. AIB will just stay with active Über PCB renditions and sticker the box’s as "advance clocks" and a new Sku's, while hopefully dropping the pricing. Or, is there some new GDDR memory that’s going to get above the 6000 MHz we see now on the GTX650?
Posted on Reply
#24
Ferrum Master
if they want war in the budget class, they should release an SKU's with unlockable cores again... it always had worked well...
Posted on Reply
#25
HumanSmoke
CasecutterWhile I wonder if they did go with Boost what's the power section going to need?
Here's a GTX 650 Ti running a 16% overclock, that "costs" all of 1 to 6 watts more.
CasecutterThere's production build-up for PCB's by the AIB's, new printing of packaging, marketing… costs?
Yup. I heard board vendors hate bringing out new models...mainly from people who have never looked at the SKU line-up's available from vendors (e.g. MSI already lists six modelsof the card, while Asus lists seven)
For what its worth, most of the GTX 650 Ti's seem to already rely on proprietary cooling solutions to differentiate themselves in the marketplace - some, like Sparkle even use a (gasp!) non reference PCB
CasecutterI think Nvidia just let loose more "higher" binned chips for less money. AIB will just stay with active Über PCB renditions and sticker the box’s as "advance clocks" and a new Sku's, while hopefully dropping the pricing. Or, is there some new GDDR memory that’s going to get above the 6000 MHz we see now on the GTX650?
Memory overclocking has already demonstrated itself to be less effective than pipeline (GPU) overclocking- both from power/heat and net gain- hence the number factory overclocked cards that sport stock clocked memory...or the fact that when AMD transformed the HD 7950 to the 7950 Boost, the memory speed was left untouched.
Posted on Reply
Add your own comment
Nov 19th, 2024 07:30 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts