Support NeoGAF

[dr_rus]

Yeah, I'm really on the fence in a lot of ways, just trying to see all my options. I'm a little worried for 1060 in the future with less DX12 and Vulkan performance and less ram than 8GB version of RX 480. Thank you for the reply!

I'm rather tired of reading this again and again here with basically no proof attached to this statement. Here, one of the worse combined results out there:

3-4% out of 30 fps is about 1 fps difference. 3-4% out of 60 fps is about 2 fps difference. You're planning on noticing such difference with a naked eye or thinking that 62 fps will be a radically better experience than 60 fps?

Also note that this cumulative percentage is built out of the following titles:
- Ashes of the Singularity, which doesn't support async compute on Pascal at the moment according to what devs said.
- Hitman, which is basically as fast in DX12 as in DX11 on AMD h/w so not really an indication of their superior DX12 performance.
- Quantum Break, which is known to favor AMD h/w in general because of a bad console code optimization.
- Rise of the Tomb Raider, which is known to favor NV h/w due to being in NV's dev program.
So that's 3 out of 4 titles which are skewed in AMD's favor by default here and only 1 title which is somewhat the opposite.

Basically, we still don't have enough data to make a call on who's doing how in DX12 or Vulkan. Considering what I've said above, I'd say that 1060 is doing fine in DX12 and there are no reasons why it would change in the future.

As for the VRAM - it's very unlikely that 6GB will ever limit the 1060, especially in 1080p.

 

[Valkyri von Thanatos]

I'm rather tired of reading this again and again here with basically no proof attached to this statement. Here, one of the worse combined results out there:

3-4% out of 30 fps is about 1 fps difference. 3-4% out of 60 fps is about 2 fps difference. You're planning on noticing such difference with a naked eye or thinking that 62 fps will be a radically better experience than 60 fps?

Also note that this cumulative percentage is built out of the following titles:
- Ashes of the Singularity, which doesn't support async compute on Pascal at the moment according to what devs said.
- Hitman, which is basically as fast in DX12 as in DX11 on AMD h/w so not really an indication of their superior DX12 performance.
- Quantum Break, which is known to favor AMD h/w in general because of a bad console code optimization.
- Rise of the Tomb Raider, which is known to favor NV h/w due to being in NV's dev program.
So that's 3 out of 4 titles which are skewed in AMD's favor by default here and only 1 title which is somewhat the opposite.

Basically, we still don't have enough data to make a call on who's doing how in DX12 or Vulkan. Considering what I've said above, I'd say that 1060 is doing fine in DX12 and there are no reasons why it would change in the future.

As for the VRAM - it's very unlikely that 6GB will ever limit the 1060, especially in 1080p.

Hmm. Looks like things aren't so cut-and-dry. Thanks for letting me know.


Edit:

Would you say then that what appears to have happened with the Kepler GPUs will not happen with Pascal, ie. the cards eventually fall behind the direct competitors they beat in the past, dues to less future conscious design choices?

 

[SmartWaffles]

Snip again

AMD features work on Nvidia cards, same as now (at least some) Nvidia features work on AMD cards as well, performance vary depending on engine and load. If you are not sure just don't bother with them, they don't really add a lot visual finesse anyway (maybe except HBAO+ but that works on AMD).

Other questions:

1. The more VRAM the merrier, go for as much as you can afford.

2. To be fair cards like a 480 or 1060 aren't meant to last long, future proof isn't the strong suit of them, cost efficiency is. In terms of API DX12 and Vulkan definitely will have a better adopt rate than DX11, so you can keep that in mind. But again, DX11 games will be the norm during both cards effective lifespan.

3. Pascal have (significantly in some cases) better async compute/DX12 capability, it is not likely to fall behind in DX12 games compared to Kepler and Maxwell which were not built with DX12 multiengine features in mind.

4. Polaris in general aren't overclockers, but neither is Pascal. By borderline capable I mean the 470 is good at 1080p60fps gaming with compromises (not everything maxed out), and it is likely to fall behind more easily than a 480. Unless it is significantly cheaper than a 480, I won't recommend it.

5. Don't go multi-GPU on mid-range cards. In fact multi-GPU in general has been a worse and worse idea.

6. TVs in the foreseeable future won't feature those tech, they are not primary PC gaming devices.

7. Warping happens if the card is really really heavy, which isn't really the case on mid-range cards. Again, backplate on those cards are mostly just fancy additions, nice to have but not necessary.

Lastly, buy ANY 480 or 1060 you can find from MSI, Asus, Gigabyte, Zotac with acceptable price and good warranty policy, you won't be disappointed with any.

 

[icecold1983]

Hmm. Looks like things aren't so cut-and-dry. Thanks for letting me know.


Edit:

Would you say then that what appears to have happened with the Kepler GPUs will not happen with Pascal, ie. the cards eventually fall behind the direct competitors they beat in the past, dues to less future conscious design choices?

It almost certainly will happen imo but to a smaller degree.

 

[spleenpoker]

Hey guys.

A bit clueless when it comes to these things but I see all these benchmarks and it seems witcher 3 gets just under 60fps on all of them with vsync obviously off but if I turn down a few settings to get it over 60 and then turn on vsync will that have a major performance hit that will just take me under 60 regardless?

 

[Marlenus]

I'm rather tired of reading this again and again here with basically no proof attached to this statement. Here, one of the worse combined results out there:

3-4% out of 30 fps is about 1 fps difference. 3-4% out of 60 fps is about 2 fps difference. You're planning on noticing such difference with a naked eye or thinking that 62 fps will be a radically better experience than 60 fps?

Also note that this cumulative percentage is built out of the following titles:
- Ashes of the Singularity, which doesn't support async compute on Pascal at the moment according to what devs said.
- Hitman, which is basically as fast in DX12 as in DX11 on AMD h/w so not really an indication of their superior DX12 performance.
- Quantum Break, which is known to favor AMD h/w in general because of a bad console code optimization.
- Rise of the Tomb Raider, which is known to favor NV h/w due to being in NV's dev program.
So that's 3 out of 4 titles which are skewed in AMD's favor by default here and only 1 title which is somewhat the opposite.

Basically, we still don't have enough data to make a call on who's doing how in DX12 or Vulkan. Considering what I've said above, I'd say that 1060 is doing fine in DX12 and there are no reasons why it would change in the future.

As for the VRAM - it's very unlikely that 6GB will ever limit the 1060, especially in 1080p.

Quoting myself from the 480 thread

Untrue. Proof.

If you compare fastest API to fastest API.

Stock 1060 wins
Ashes
Tomb Raider

Stock 480 wins
Doom
Forza
Hitman
Quantum Break
Gears

They tie in
Total war: warhammer.

The 1060 has performance regressions in
Ashes
Doom
Hitman
Total War: Warhammer
Tomb Raider

The 480 has regressions in
Tomb Raider

To me it looks like the 480 is a safer bet if you intend to keep the card for more than a year. It performs better in most games with a low level API and it also gains performance when using the low level API.

 

[dex3108]

none that i've seen, but i own that card if you have any questions.

I am mostly interested in temps and noise. Also did you try to overclock it? Tnx.

 

[Xyber]

Quoting myself from the 480 thread

I wouldn't say that that is the best way to compare them though. Nvidia having a bit worse performance in DX12 than in 11 in some games is not very surprising to me. It's a very new api where the developers actually have to put in some work to optimize it.

AMD on the other hand had less than great DX11 drivers and if the games run worse in DX12 then the developers have messed up.

Nvidia never gain as much as AMD in DX12 because their DX11 drivers were already pretty damn good when it came to reducing the overhead. AMD has always had hardware on par (or even better than Nvidia) but their DX11 drivers have held them back badly, which these new api's have shown.

 

[Durante]

Quoting myself from the 480 thread

It's much easier to regress from a very good level (NV's DX11 drivers) than from a mediocre level (AMD's DX11 drivers [and let's just not talk about their OpenGL drivers]).

 

[ISee]

Hey guys.

A bit clueless when it comes to these things but I see all these benchmarks and it seems witcher 3 gets just under 60fps on all of them with vsync obviously off but if I turn down a few settings to get it over 60 and then turn on vsync will that have a major performance hit that will just take me under 60 regardless?

you'll be fine.

4059 - Time Spy http://www.3dmark.com/3dm/13685930

8004 - Fire Strike http://www.3dmark.com/3dm/13687044

Thank you.
1060 async compute really boosts the 1060 to 980 levels in Timespy. Very impressive.

Had to lower the OC for Fire Strike since I kept getting driver crashes. They weren't exaggerating about it being a demanding benchmark.

It surely is and especially Timespy seems like a good indication for intestable overclocks. I even had to readjust voltage on my 980 bios just because of it (+6mv).

 

[Marlenus]

It's much easier to regress from a very good level (NV's DX11 drivers) than from a mediocre level (AMD's DX11 drivers [and let's just not talk about their OpenGL drivers]).

This is true but misses the point I was trying to get at.

When there is a DX12 option the 480 is faster than the 1060 in the majority of titles so far. It is also faster in all the UWP titles tested. Based on the current trend it seems more likely that in a year or so when there are even more titles with DX12 or Vulcan support and more UWP titles the 480 will hold up better than the 1060.

 

[SomedayTheFire]

I am mostly interested in temps and noise. Also did you try to overclock it? Tnx.

haven't tried to overclock it yet. noise is pretty much non existent a huge huge upgrade on my last card. don't hear the fans until they're nearly 70% speed and haven't had the fans go above 50% in real time gameplay. temp's are usually around 58-60C in gameplay with the highest I've gotten being 74C in a benchmark.

 

[blastprocessor]

When there is a DX12 option the 480 is faster than the 1060 in the majority of titles so far. It is also faster in all the UWP titles tested. Based on the current trend it seems more likely that in a year or so when there are even more titles with DX12 or Vulcan support and more UWP titles the 480 will hold up better than the 1060.

There's no doubt Vulkan performance on 1060 is really shitty right now. I hope Nvidia close the gap but hey it's only one title so far.

 

[Durante]

There's no doubt Vulkan performance on 1060 is really shitty right now. I hope Nvidia close the gap but hey it's only one title so far.

"Vulkan" performance is not bad at all. Really, equating Doom performance with Vulkan performance makes no sense. We currently have 2 games with unfinished/incomplete Vulkan renderers available: Doom and The Talos Principle. In one of them one card does much better, in the other one the other does much better. Neither of them is final or should be taken as a reference for all future Vulkan performance.

(Actually, there's a third game with a Vulkan renderer, DOTA 2, but I couldn't find any benchmarks of that with these cards -- and it has such low requirements that it's hard to isolate CPU factors)

 

[blastprocessor]

"Vulkan" performance is not bad at all. Really, equating Doom performance with Vulkan performance makes no sense.

That's why l said it's only one title so far.

 

[spleenpoker]

you'll be fine

Sorry but can you elaborate on that? I'm on the edge of buying one as we speak but would like to squish my doubts before I purchase

This is only a stop gap graphics card for about a year or two till we see nvidias new architecture and AMDs offerings

 

[ISee]

Sorry but can you elaborate on that? I'm on the edge of buying one as we speak but would like to squish my doubts before I purchase

This is only a stop gap graphics card for about a year or two till we see nvidias new architecture and AMDs offerings

Vsync takes some performance, but it's not really a deal breaker. If you're able to reach e.g. 65+ fps you won't drop below 60 just because of vsync.

 

[bj00rn_]

One thing my ignorant mind doesn't quite understand is; The majority of people have Nvidia GPUs, and the best performance is on DX11 (and most games), why all this focus on DX12 and Vulkan right now? Why compare Vulkan benchmarks when it's slower than DX11 (yet?)?

 

[dr_rus]

Would you say then that what appears to have happened with the Kepler GPUs will not happen with Pascal, ie. the cards eventually fall behind the direct competitors they beat in the past, dues to less future conscious design choices?

It won't, for several reasons, one being that Pascal is actually ahead of GCN in features (was the opposite with Kepler) and another being that Pascal don't really have anything which may slow it down considerably when running GCN code. Some performance degradation is possible but it will most likely be limited to bad console ports or AMD sponsored titles.

Quoting myself from the 480 thread

As I've said there, you're using just one review and instead of looking at the numbers it provides you're just saying "it's faster". While even there it's not as clear cut as you want to make it look, for example, "Stock 480 wins Quantum Break":

Then you should weigh what title is sponsored by whom into the mix and preferably remove them from comparison.

Then you should actually consider that comparing stock cards are meaningless as reference 480 isn't selling too well and reference 1060 is close to being unavailable.

So it's not a clear cut at all.

 

[Knch]

So 1060 will never close to DX12 and Vulkan gap because it just lacks the hardware chops, is that right?

Pascal handles async in a different way than previous Nvidia architectures (from what I can remember of Nvidia's slides) and probably has some room for optimisation in their drivers. If async is a big thing (and it "should" be) then I would rather bet on AMD's approach of more and slower cores than Nvidia's fewer and faster cores.

Would you say the 470 could feasibly clock to nearly (a couple %s difference) from 480, or is that reaching? Do you think the 470 will have less supply issues than the 480? I'm worried because I want to get everything installed before Deus Ex MD releases on August 23rd, and I only have a couple weeks to enjoy that game during my break before my fall term at college. I'm trying to get something going by August 20th at the latest and I'm worried the 470 will take too long to come out.

The difference in core count is quite small and it's the same architectures. The problem is the slower memory chips, which might not OC that well. If it weren't for that I'd be quite confident in saying yes, as it is, I'm not quite sure.

No idea about availability, but this shouldn't be a paper launch.

I will check out ULMB. (That's "ultra low motion blur" right?)

Right, it's basically strobing a backlight to reduce motion blur, which I'm not a fan of. (I prefer an IPS @ 75Hz max)

Is warping dangerous? What exactly is it? Could it kill or weaken a card after a few years?

Warping is the PCB actually deforming, which is highly unlikely as most coolers are connected at multiple points across the entire length of the cards these days.

If the backplate is aluminium and has thermal transfer material applied, it very slightly will improve thermals (assuming your case has airflow across it.) Otherwise, it's purely for aesthetic purposes or for the insanely paranoid.

Lastly, are there any particular custom cards (480 or 1060) that you would recommend?

Just go for whichever you fancy most, the differences are too minor to actually recommend one over the other. Bigger heatsink/Fans = less noise, that and whoever is cheapest will be my criteria for choosing.

 

[BlazinSkatinAxel]

Are all sellers currently sold out?

 

[coopolon]

My local microcenter has the zotac gtx 1060 AMP! edition for $300 in stock. Is this a good version of the card and is that a good price?

It seems like everywhere I check online the companies I'm more familiar with (gigabyte, EVGA, MSI) are out of stock and I kind of a hankering to get this card this weekend.

Edit: Well, I guess I already know newegg has same card in stock for $30 less so the price isn't great.

 

[Meh3D]

Any good write ups on the minis? I've had my eye on the EVGA one for a small build I want to gift to my brother.

Do these fare better than those 970 minis?

(Surprised there are no 480 minis)

 

[Dead Prince]

anyone tried with GW2? all i see is green and warning says my card isn't supported.

 

[ambar_hitman]

I'm rather tired of reading this again and again here with basically no proof attached to this statement. Here, one of the worse combined results out there:

3-4% out of 30 fps is about 1 fps difference. 3-4% out of 60 fps is about 2 fps difference. You're planning on noticing such difference with a naked eye or thinking that 62 fps will be a radically better experience than 60 fps?

Also note that this cumulative percentage is built out of the following titles:
- Ashes of the Singularity, which doesn't support async compute on Pascal at the moment according to what devs said.
- Hitman, which is basically as fast in DX12 as in DX11 on AMD h/w so not really an indication of their superior DX12 performance.
- Quantum Break, which is known to favor AMD h/w in general because of a bad console code optimization.
- Rise of the Tomb Raider, which is known to favor NV h/w due to being in NV's dev program.
So that's 3 out of 4 titles which are skewed in AMD's favor by default here and only 1 title which is somewhat the opposite.

Basically, we still don't have enough data to make a call on who's doing how in DX12 or Vulkan. Considering what I've said above, I'd say that 1060 is doing fine in DX12 and there are no reasons why it would change in the future.

As for the VRAM - it's very unlikely that 6GB will ever limit the 1060, especially in 1080p.

And then there is this:

Found this interesting video.

Vulkan on older CPUs reverses the trend. 1060 beats 480 handsomely:

https://youtu.be/pjZ3MmxP0Fg?t=4m33s

I have AMD 1055T. My current R9 280 blows Doom away at 1080p with Vulkan.
Let's see how 1060 does with my CPU.

Exactly reverse results with older CPUs. Vulkan is very confusing.

 

[Smidget]

Got my EVGA 1060 SC from Newegg today. Did the Guru3D GPU Uninstaller, swap, and then new drivers and everything seems to be working swimmingly.

Have a i5 2500K O/Ced: http://www.3dmark.com/3dm/13712367

Assuming those are standard results?

 

[icecold1983]

It won't, for several reasons, one being that Pascal is actually ahead of GCN in features (was the opposite with Kepler) and another being that Pascal don't really have anything which may slow it down considerably when running GCN code. Some performance degradation is possible but it will most likely be limited to bad console ports or AMD sponsored titles.



As I've said there, you're using just one review and instead of looking at the numbers it provides you're just saying "it's faster". While even there it's not as clear cut as you want to make it look, for example, "Stock 480 wins Quantum Break":

Then you should weigh what title is sponsored by whom into the mix and preferably remove them from comparison.

Then you should actually consider that comparing stock cards are meaningless as reference 480 isn't selling too well and reference 1060 is close to being unavailable.

So it's not a clear cut at all.

nvidia performs much better when you leave god rays at medium. turning them to ultra is what kills them

 

[jonno394]

Seeing as i've just installed a new CPU cooler and my Gigabyte 1060, I can't be too sure what was so loud before hand while playing Talos Principle, but everything is amazingly quiet atm.

When yesterday the fans would be loud enough to hear from my sofa, everything runs so quiet now. The cards a great looker, not really a techy so can't comment on how good it is myself, but I'm happy with my purchase

 

[slash000]

Seeing as i've just installed a new CPU cooler and my Gigabyte 1060, I can't be too sure what was so loud before hand while playing Talos Principle, but everything is amazingly quiet atm.

When yesterday the fans would be loud enough to hear from my sofa, everything runs so quiet now. The cards a great looker, not really a techy so can't comment on how good it is myself, but I'm happy with my purchase

What CPU cooler, may I ask?
I'm about to do the same -- replace the noisy crap in my computer, sounds like a jet engine atm (awaiting my 1060 to come in soon, want to replace my case fans and CPU hsf in the meantime)

 

[jonno394]

What CPU cooler, may I ask?
I'm about to do the same -- replace the noisy crap in my computer, sounds like a jet engine atm (awaiting my 1060 to come in soon, want to replace my case fans and CPU hsf in the meantime)

I stuck with a cheap one but everyone seems to recommend it, Cooler Master 212x. Fiddly to attach though :/

 

[terrible]

And then there is this:



Exactly reverse results with older CPUs. Vulkan is very confusing.

It's unfortunate that more places aren't doing benchmarks with older CPUs since you'd think a large amount of people upgrading to these lower end cards would have older CPUs. I know I do (2500K).

 

[ambar_hitman]

Interesting video:

https://www.youtube.com/watch?v=Xuea5nt4TnY

ASYNC Off or ON in Doom VULKAN doesn't make much difference with 480 FPS.

 

[Anteater]

I just went out and picked up an Asus 1060 Dual since I saw one in a local store (I have the choice of Gigabyte but this one was cheaper, and I couldn't ask Gaf since my internet crapped out when I was right outside the store lol).

Replaced my 760 GTX and my computer still turns on so I guess I didn't break anything, everytime I open up my computer I feel like I'm doing a fucking surgery.

Are the Asus softwares necessary? Like GPU Tweak II or whatever.

 

[Frankfurter]

It's unfortunate that more places aren't doing benchmarks with older CPUs since you'd think a large amount of people upgrading to these lower end cards would have older CPUs. I know I do (2500K).

You can always check computerbase.de for that (in German, but I guess you'll get the benchmark bars, numbers etc.). They always have a special test using the 2500k (@stock) with a few games.

See for the 1060: https://www.computerbase.de/2016-07...diagramm-xcom-2-1920-1080-intel-core-i5-2500k

 

[ISee]

I just went out and picked up an Asus 1060 Dual since I saw one in a local store (I have the choice of Gigabyte but this one was cheaper, and I couldn't ask Gaf since my internet crapped out when I was right outside the store lol).

Replaced my 760 GTX and my computer still turns on so I guess I didn't break anything, everytime I open up my computer I feel like I'm doing a fucking surgery.

Are the Asus softwares necessary? Like GPU Tweak II or whatever.

Nope, if you don't want to overclock you're free to uninstall it.

 

[terrible]

You can always check computerbase.de for that (in German, but I guess you'll get the benchmark bars, numbers etc.). They always have a special test using the 2500k (@stock) with a few games.

See for the 1060: https://www.computerbase.de/2016-07...diagramm-xcom-2-1920-1080-intel-core-i5-2500k

Interesting, thanks. I wonder what a 4.5ghz 2500K would do against the 6700K in those benchmarks.

 

[Anteater]

Nope, if you don't want to overclock you're free to uninstall it.

ooh so that's what it is, thanks.

I think I can OC with afterburner, I'll just use that so I don't have another program running in the background

 

[dr_rus]

nvidia performs much better when you leave god rays at medium. turning them to ultra is what kills them

Which basically means that however the god rays are implemented there isn't optimal for NV GPUs. Highly unlikely that it has anything to do with DX12 in general. From what they've disclosed on their volumetric lighting implementation it seems that they've decided against using, for example, volumetric sparse textures due to concerns on h/w support (strange decision considering the game is console/DX12 exclusive and all DX12 h/w should have such support) and opted for a s/w (GPU compute) based adaptive voxel tree solution instead. As a result, the GPUs which have more pure compute power wins over those which have less actual flops. Async compute gains help AMD h/w a bit here most likely but it's still secondary to the real reason which is pure flops advantage.

 

[icecold1983]

Which basically means that however the god rays are implemented there isn't optimal for NV GPUs. Highly unlikely that it has anything to do with DX12 in general. From what they've disclosed on their volumetric lighting implementation it seems that they've decided against using, for example, volumetric sparse textures due to concerns on h/w support (strange decision considering the game is console/DX12 exclusive and all DX12 h/w should have such support) and opted for a s/w (GPU compute) based adaptive voxel tree solution instead. As a result, the GPUs which have more pure compute power wins over those which have less actual flops. Async compute gains help AMD h/w a bit here most likely but it's still secondary to the real reason which is pure flops advantage.

The reason most of us think nvidia will do poorly in dx12 is because of software. Amd will be fine because a lot of the work done on consoles will have some nice carryover. I dont see many devs doing all the work necessary to get optimal perf on nvidia. Async comoute is just a nice bonus on top. Shader instristics could be huge tho if it gains traction

 

[dr_rus]

The reason most of us think nvidia will do poorly in dx12 is because of software. Amd will be fine because a lot of the work done on consoles will have some nice carryover. I dont see many devs doing all the work necessary to get optimal perf on nvidia. Async comoute is just a nice bonus on top. Shader instristics could be huge tho if it gains traction

But this has nothing to do with DX12 as this will equally appear on any API out there, and no, "most of people" thinking this is actually thinking of some magical advantage of GCN h/w in DX12.

I think that most devs will in fact do the work to get optimal performance on NV. Simply because this is how it was, is and there's no reason why it should change. The cases where such optimizations are missing can be counted with one hand today, after almost three years of new consoles being on the market.

Pascal gain performance from async compute too, shader intrinsics are supported by NV since ages ago and what they allow will likely become a standard part in DX SM 6.0. PC h/w isn't standing still waiting for AMD to command what to do next. Pascal is basically already more efficient at running GCN console code than the modern GCN itself, and if AMD is planning to be "fine" by doing nothing then they may be in big trouble by the time Volta hits the retail.

 

[Alchemy]

It's unfortunate that more places aren't doing benchmarks with older CPUs since you'd think a large amount of people upgrading to these lower end cards would have older CPUs. I know I do (2500K).

I still have an i7 860... so yeah. Still really happy with this GPU upgrade though.

 

[Ruff]

My GPU has finally crapped out on me, convenient since I was looking at this for an upgrade anyway. Only store that has any stock is only has some left by PALIT, anyone have any experience with this manufacturer?

 

[coopolon]

Microcenter had the pny for regular MSRP so went ahead with that. Pretty pleased so far, I finally get to have in home streaming just work like magic.

I think the rx480 is probably a better deal performance wise, I just can't take any more shitty drivers from amd, I've finally learned I just need to pay the Nvidia tax.

 

[DMTripper]

What do the UK guys think of the current 1060 pricing? Just looking at overclockers and they seem to range from £270-325... Is this legit?

 

[Delusibeta]

What do the UK guys think of the current 1060 pricing? Just looking at overclockers and they seem to range from £270-325... Is this legit?

Considering I haven't found a non-Overclockers 1060 for less than £290, I guess that's the best buys so far.

[Edit] Above sentence is wrong: Amazon has 1060s for £240.

 

[blastprocessor]

What do the UK guys think of the current 1060 pricing? Just looking at overclockers and they seem to range from £270-325... Is this legit?

Scan computers has them for £249 and in stock at that price

 

[dr_rus]

Palit GTX 1060 Super JetStream 6 GB

 

[MichaelEM3]

Is it just me or has this card not been in-stock on Amazon or Newegg in the US for the last 3 days? Nothing around $250, at least.

 

[Bullza2o]

Is it just me or has this card not been in-stock on Amazon or Newegg in the US for the last 3 days? Nothing around $250, at least.

I've been waiting for my EVGA 1060SC since July 20... Amazon is still not giving any shipping dates.

 

[super-famicom]

Is it just me or has this card not been in-stock on Amazon or Newegg in the US for the last 3 days? Nothing around $250, at least.

Just check out nowinstock on a regular basis. It's how I was able to get my EVGA SC 1060 last week. As of this post, the Asus Strix and Gigabyte cards are in stock at Newegg and B&H Photo.

http://www.nowinstock.net/computers/videocards/nvidia/gtx1060/