• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Geforce 750 Ti - Maxwell!?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Its Kepler, not Maxwell.

Device 10de:11c7
Name: GK106 [GeForce GTX 750 Ti]

That was kinda my point...

Are we sure that GTX 750 Ti is Maxwell?
It sounds weird putting Maxwell in the 700 series along with other Kepler`s.
The specs seems like something Nvidia would do to rebrands: Overclock them to 1-1.1GHz and call it a day.
Maxwell in 28nm? Why? Both Fermi and Kepler both launched with new architecture and new node.
 
Last edited:
It sound really weird having this be Maxwell. Surely Maxwell implies 20nm TSMC as well as a new architecture in the way Nvidia usually works?
 
It sound really weird having this be Maxwell. Surely Maxwell implies 20nm TSMC as well as a new architecture in the way Nvidia usually works?

My theory was that nVidia backported Maxwell to 28 nm when they realized there was going to be trouble getting to TSMC 20 nm.
 
Here is the rumored spec for GTX 750 TI
NVIDIA-GeForce-GTX-750-Ti-GPUz.jpg


Its from early september last year

and it looks awefully like this Kepler mobile GPU

iMac675mx_zps48409bf5.gif



But the specs for GTX 750 Ti could be fake. I don`t know.
:hmm:

Yeah it does, although gpu-z seems to not show correct info for GTX675MX, should be 960/80/32 shaders/tmu/rops. It also looks a lot like GTX660 but 32 instead of 24 rops and that would make more sense to me, using GK106 instead of cutdown GK104.

Not sure where the Maxwell rumor comes from. Seems like a normal move to me: 650Ti Boost vs 7850, now 750Ti vs R9 270.
 
Yeah it does, although gpu-z seems to not show correct info for GTX675MX, should be 960/80/32 shaders/tmu/rops. It also looks a lot like GTX660 but 32 instead of 24 rops and that would make more sense to me, using GK106 instead of cutdown GK104.

Not sure where the Maxwell rumor comes from. Seems like a normal move to me: 650Ti Boost vs 7850, now 750Ti vs R9 270.

Going GK106 instead of GK104 means 192bit bus instead of 256bit. 675MX is GK104.

The Maxwell rumor all comes from Sweclockers. Which is known to be reliable, but I guess they miss their targets sometimes too.
Not sure if it truly is a rebrand or not though. The specs could be wrong too. Confusing.
 
Last edited:
It performs pretty impressively, but for some reason the 750 and 750Ti have a 6 pin connector, I thought they would use less power than that.

You know that there is a wide range of power to be consumed between no additional connector and one 6 pin, right? TDP of the 650 Ti is 110W, PCIe only provides 75W iirc. So if the 750 Ti consumes 90W and is 30% faster than the 650 Ti, that would still be a seizable step in perf/W and at the same time explain the connector. Maybe the 750 uses the same board and the card was a pre-production sample. I would be puzzled too, if the 750 were to have a 6-pin. For the 750 Ti it is plausible though, especially if it's on 28nm still which I strongly assume.
 
Going GK106 instead of GK104 means 192bit bus instead of 256bit. 675MX is GK104.

The Maxwell rumor all comes from Sweclockers. Which is known to be reliable, but I guess they miss their targets sometimes too.
Not sure if it truly is a rebrand or not though. The specs could be wrong too. Confusing.

You're right, I missed that mem bus. So maybe Nvidia has a big pile of bad GK104 chips by now. 760 SE seems like are more appropriate name then, in line with Nvidia conventions.
 
Back
Top