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

NVIDIA Rushes in GTX 1060 with GDDR5X to Counter AMD Radeon RX 590 Threat

btarunr

Editor & Senior Moderator
Staff member
Joined
Oct 9, 2007
Messages
47,670 (7.43/day)
Location
Dublin, Ireland
System Name RBMK-1000
Processor AMD Ryzen 7 5700G
Motherboard Gigabyte B550 AORUS Elite V2
Cooling DeepCool Gammax L240 V2
Memory 2x 16GB DDR4-3200
Video Card(s) Galax RTX 4070 Ti EX
Storage Samsung 990 1TB
Display(s) BenQ 1440p 60 Hz 27-inch
Case Corsair Carbide 100R
Audio Device(s) ASUS SupremeFX S1220A
Power Supply Cooler Master MWE Gold 650W
Mouse ASUS ROG Strix Impact
Keyboard Gamdias Hermes E2
Software Windows 11 Pro
AMD is giving final touches to its Radeon RX 590 graphics card, which is rumored to be based on an efficient new rendition of the "Polaris" silicon, which could disturb NVIDIA's product lineup between the GTX 1060 series and the GTX 1070, as its new RTX 2060 series is nowhere in sight. In a bid to thwart this threat, NVIDIA is preparing a variant of the GeForce GTX 1060 with faster GDDR5X memory.

The current GTX 1060 6 GB is endowed with 8 Gbps GDDR5 memory, which at its 192-bit bus width works out to a memory bandwidth of 192 GB/s. NVIDIA had attempted to improve its competitive position once, by creating a shortlived sub-variant of this SKU with 9 Gbps GDDR5 memory (211 GB/s). Switching to 10 Gbps GDDR5X memory would give the chip 240 GB/s memory bandwidth, and 11 Gbps (unlikely because expensive), would yield 264 GB/s. With the GP106 silicon maxed out, it's also possible the new GTX 1060 could be based on a heavily cut down GP104, possibly even with 192-bit memory, which explains GDDR5X memory.



View at TechPowerUp Main Site
 
Muddying up the product stack again...
 
NVIDIA has a clear chance to rebrand such card. Probably not RTX\GTX 2060, but possibly a 2050 Ti moniker.
I also wouldn't be surprised if this product will end up with a 299$ price tag
 
Would it make sense, though? I'm not sure GP106 is that bandwidth limited and it's not like Pascal is the most bandwidth-hungry GPU architecture ever made.
 
I'd rather see the 1070 get a price cut instead.

Would it make sense, though? I'm not sure GP106 is that bandwidth limited, especially considering Pascal isn't the most bandwidth hungry GPU architecture ever made.

Yeah, that is kind of what I'm thinking too. The switch from 8Gbps to 9Gbps didn't really make much of a difference, it just made the cards more expensive. Which is why the 9Gbps version was short lived.

So I can't see going from 8 to 10Gbps making a big difference either, again other than price that is.
 
But why...
It's still a gp106, you can give it GDDR8, it wont make much of a difference.
 
Riight, so we buy this old tatt, lol jk ..
Only to have a 2060 in January, only someone with a dire need should contemplate this nonsense imho.
 
Muddying up the product stack again...
Yeah, I recall they did this with the 460 different memory amounts and speed but no designation, and also the 260 with 2 different variants of the core. :wtf:
 
Yeah, I recall they did this with the 460 different memory amounts and speed but no designation, and also the 260 with 2 different variants of the core. :wtf:
You don't need to go that far into the past.
1060 6GB and 3GB have different amount of cuda cores. You buy it and think you only made a VRAM sacrifice, but surprise, surprise...
 
You don't need to go that far into the past.
1060 6GB and 3GB have different amount of cuda cores. You buy it and think you only made a VRAM sacrifice, but surprise, surprise...
Exactly this and now theyll be GDDR5 and GDDR5X cards to add further confusion.
Edit: wrong RAM...
 
Last edited:
Exactly this and now theyll be GDDR5 and GDDR6 cards to add further confusion.
I don't think we'll ever see GDDR6 on Pascal, the architecture doesn't even support it.
 
You don't need to go that far into the past.
1060 6GB and 3GB have different amount of cuda cores. You buy it and think you only made a VRAM sacrifice, but surprise, surprise...

Also the 5GB OEM version.

This is downright silly. Drop the price of the current 1060s and the 1070 and be done with it.
 
You don't need to go that far into the past.
1060 6GB and 3GB have different amount of cuda cores. You buy it and think you only made a VRAM sacrifice, but surprise, surprise...

At least nVidia always makes some way to tell the difference between the different cards. Unlike a certain other graphics card manufacturer that will remain nameless...
 
I don’t see the point of this at all

Better pricing would be way more competitive than releasing yet another GP106 variant that would only have a larger premium
 
This whole thing is assenine. Now we will have 1060’s with GDDR5X and the much better 1070 with GDDR5.

I don’t see the point of this at all

Better pricing would be way more competitive than releasing yet another GP106 variant that would only have a larger premium
I think they must have a surplus of GP106 left and no cards to put them on yet.
 
This whole thing is assenine. Now we will have 1060’s with GDDR5X and the much better 1070 with GDDR5.


I think they must have a surplus of GP106 left and no cards to put them on yet.
Make a refresh called GTX 2050 with 6GB VRAM GDDR5 with good clocks at around $199 and Nvidia will have a winner on their hands for the lower end segment

Pricing is always key for consumers in the low-mid range section
 
I don't think we'll ever see GDDR6 on Pascal, the architecture doesn't even support it.
Sorry typo I meant GDDR5X
 
At least nVidia always makes some way to tell the difference between the different cards. Unlike a certain other graphics card manufacturer that will remain nameless...

Yep. Can't imagine how many people purchased 560's with the lower core count with the only nomenclature change being a cryptic part number difference that may or may not mean it was actually a locked down card.

Or a 550 with an arbitrary number of shaders that fell between the 550 and 560 based on the 560 core.

I remember the old NV gtx 260 core 216 and they bragged that up all over the box. 1070 gained a "ti" model... Lol what a joke from the other side.
 
Yep. Can't imagine how many people purchased 560's with the lower core count with the only nomenclature change being a cryptic part number difference that may or may not mean it was actually a locked down card.

Or a 550 with an arbitrary number of shaders that fell between the 550 and 560 based on the 560 core.

I remember the old NV gtx 260 core 216 and they bragged that up all over the box. 1070 gained a "ti" model... Lol what a joke from the other side.
Don’t forget the 460’s. No designation between different memory models unless you looked at the specs. If I recall there was 768 mb and 1.5GB.
 
and also the 260 with 2 different variants of the core.
There were three actually, the 65nm & 55nm 192 SP models and the 55nm 216 SP model.
 
At least nVidia always makes some way to tell the difference between the different cards.
Is that so.
Usually when you buy a card that has different amount of memory, you assume the GPU is the same. Like it has been for years before. Example RX480. Both 8GB and 4GB have the same chip underneath.
MX150 has 2 variants (or was it the mx130 hmmm). One is the default (being advertised as such) and its, in name, indistinguishable slower brother, pushed into the market in silence. You wont know which one you are getting until you buy the laptop and run GPUZ, at which point it's too late.

As for AMD side of story, we have the 550 with different specs in the same name and I believe a 460. The new 580 for the Chinese market has it's core count in its name on the Chinese website.

So please explain to me how is one company better than the other and vice versa, considering they do mostly the same stuff.
 
As for AMD side of story, we have the 550 with different specs in the same name and I believe a 460. The new 580 for the Chinese market has it's core count in its name on the Chinese website.

The entire bottom half of their product stack is arbitrary names that don't line up.

Don’t forget the 460’s. No designation between different memory models unless you looked at the specs. If I recall there was 768 mb and 1.5GB.

The RX 460? That was only offered in one core count unless you count the MacBook version. The memory difference was 2/4gb and they were labeled as such. I would say that was less misleading than the 1060 3/6gb with its varied core count.

The gtx 460 was a bit funky v1 was 336 cores and 256 bit the v2 was 336 cores and 192 bit and the SE was 288 cores and 256 bit plus the 768mb model that was 336 cores and 192 bit lol.
 
It would require a new die with a different memory controller. Unlikely Nvidia would bother with that.

Or they could use a severly cut down GP104 . That's not as crazy as it sounds , that is after all nothing more than a mid-range GPU, size wise. Regardless, how many different 1060s are there now ? I lost count, it's ridiculous.

Pascal is the most bandwidth-hungry GPU architecture ever made.

Every GPU architecture is hungry for more memory bandwidth to a degree.
 
Last edited:
the rumor is this is a 1060 6 GB with GP104 on a 1080 PCB

here is the entire stack of cards
Untitled.jpg
 
Last edited:
Back
Top