Wednesday, December 5th 2018

NVIDIA Unveils GeForce GTX 1070 with GDDR5X Memory

It looks like NVIDIA bought itself a mountain of unsold GDDR5X memory chips, and is now refreshing its own mountain of unsold GP104 inventory, to make products more presentable to consumers in the wake of its RTX 20-series and real-time ray-tracing lure. First, it was the GP104-based GTX 1060 6 GB with GDDR5X memory, and now it's the significantly faster GeForce GTX 1070, which is receiving the newer memory, along with otherwise unchanged specifications. ZOTAC is among the first NVIDIA add-in card partners ready with one such cards, the GTX 1070 AMP Extreme Core GDDR5X (model: ZT-P10700Q-10P).

Much like the GTX 1060 6 GB GDDR5X, this otherwise factory-overclocked ZOTAC card sticks to a memory clock speed of 8.00 GHz, despite using GDDR5X memory chips that are rated for 10 Gbps. It features 8 GB of it across the chip's full 256-bit memory bus width. The GPU is factory-overclocked by ZOTAC to tick at 1607 MHz, with 1797 MHz GPU Boost, which are below the clock-speeds of the GDDR5 AMP Extreme SKU, that has not just higher 1805 MHz GPU Boost frequency, but also overclocked memory at 8.20 GHz. Out of the box, this card's performance shouldn't be distinguishable from the GDDR5 AMP Core, but the memory alone should serve up a significant overclocking headroom.
Add your own comment

78 Comments on NVIDIA Unveils GeForce GTX 1070 with GDDR5X Memory

#26
Turmania
This is getting ridiculous both Nvdia and AMD in GPU sector.I really hope Intel succeeds and be competitive in GPU sector or we are left with these two making mockery of us consumers.not that Intel is a saint but competition is good.
Posted on Reply
#27
The Quim Reaper
AssimilatorWhich brings up another interesting question: are the prices of Turing so high only because NVIDIA is greedy, or was this writedown cost already factored into them from the start?
This is Nvidia, It wont be an either/or choice. They'll be looking to make more money from both ends.
Posted on Reply
#28
silentbogo
I'm losing my hope for humanity. We get a decent bump in memory throughput, and people rant about it like it's a bad thing.
Posted on Reply
#29
delshay
the54thvoidYup. Been happy with mine since day one.

And as for using better memory on an old card, can someone explain why that is a bad thing? Look, here is item 'A' which we now sell with a better component. As long as price doesn't inflate.
Using faster memory is a win, win for the end user(s). You have a choice to either up the clock or keep the same clock & lower the timings. Either way the GPU core will perform better.
Posted on Reply
#30
R0H1T
silentbogoI'm losing my hope for humanity. We get a decent bump in memory throughput, and people rant about it like it's a bad thing.
This isn't a choice, for Nvidia or Micron. They have to use the GDDR5x, otherwise someone somewhere will have to write off a lot of unused inventory.
Sure you could argue about the mining boom/bust but this situation was kinda avoidable & if it were avoided, who knows we might've seen better prices for RTX.
Posted on Reply
#31
[XC] Oj101
R0H1TSource for the 1080 boards being used in GTX 1060 or 1070 :confused:
If you look at the custom cards they have the SLI fingers at the top, some with a sticker over them saying that SLI is not supported. If you remove the heat sink you see that the board is identical in more ways than that.
Posted on Reply
#32
silentbogo
R0H1TThis isn't a choice, for Nvidia or Micron.
It does not really matter whether it's a choice or not, and what were the circumstances. The end result is an upgrade with either small or no price increase for the end-user.
The same story happened on few occasions, like with several revisions of GT730, or long-forgotten GTX650Ti [boost], or the recent GTX1060 but in all cases people find justification for complaints.

If it was something along the lines of GT1030 DDR4, or GTX970 misinformation, or recent shenanigans with an entire Polaris refresh lineup, then it's worth complaining about.
Posted on Reply
#33
m4dn355
This is getting out of control. . .
Posted on Reply
#34
Vayra86
silentbogoIt does not really matter whether it's a choice or not, and what were the circumstances. The end result is an upgrade with either small or no price increase for the end-user.
The same story happened on few occasions, like with several revisions of GT730, or long-forgotten GTX650Ti [boost], or the recent GTX1060 but in all cases people find justification for complaints.

If it was something along the lines of GT1030 DDR4, or GTX970 misinformation, or recent shenanigans with an entire Polaris refresh lineup, then it's worth complaining about.
I think the pain is in the overall lack of progress, and these refreshes are just examples of it. Yes its free performance, in a rare edge case, maybe you will win 2% and in most others its completely irrelevant (remember the 1070ti running on GDDR5...). Its also the same price/perf as it was 3 years ago.

650ti boost for example was launched in a very lively GPU landscape, closely followed by a Kepler Refresh that pushed the GTX 680 to a lower price point with faster memory. At the same time, the 650ti boost performed like a 660 but at the cost of a 650ti. These GDDR5X GPUs won't be making any sort of jump like that, and they are visibly handicapped to make them slower. If Nvidia had clocked the memory at its stock 10Gb I think the noise would be a whole lot different.

Right now you can see them putting effort in handicapping performance, almost 3 years after the fact... That is something else and if there is one thing it isn't, its 'for the gamers'. Being the optimist in saying 'but its free performance' is... well, being overly optimistic in my opinion. This performance isn't free at all, the price is too high to begin with. Especially when you know these chips and boards could have become 1080's that compete with a more expensive RTX alternative.
Posted on Reply
#35
jabbadap
Well I'm more interested to see what gpu chip it has. Is it mining gpu GP104-100-A1 or vanilla gtx1070 gpu GP104-200-A1. Funny thing with NVIDIA P104-100 mining card is that it has 8x8Gb 10Gbps GDDR5X memory chips but 4GB memory by bios. If one could get these biosses to those mining cards, would they work as normal video cards with full memory?
Posted on Reply
#36
ExV6k
ArbitraryAffectionPoopscal with faster memory is still Poopscal. Except the memory isn't even faster out of the box. Huh.
Huh, someone's mad. Does it bother you so much that Nvidia are this successful?
Posted on Reply
#37
jabbadap
silentbogoI'm losing my hope for humanity. We get a decent bump in memory throughput, and people rant about it like it's a bad thing.
Not to mention news piece says it's under clocked to 8Gbps, so it's changes nothing on stock. But the the memory OC potential should be massive on this.
Posted on Reply
#38
siluro818
Vayra86Another issue with Vega is that a lot of its AIB versions are hit or miss. If you don't have one of the great ones, you have a shitty one. And guess what, the nice ones do cost more. Then factor in the power draw gap across a few years of light gaming and *poof* price difference gone.
Not sure if this is really the case with Vega AIBs to be honest. The Sapphire 64 is undeniably the best build there is, as is usually the case with their AMD cards, but there is essentially no difference whether you'll get a Pulse or Nitro 56 in terms of performance and OC or price for that matter.

Meanwhile the Gigabyte 56 is being offered for 10% less than the Sapphire cards, but again there's no difference in performance whatsoever, while currently their 64 is the same price as a Nitro, with obviously worse build. Yet, while it has worse quality cooling, mine GB64 (card size was a priority when I bought it) runs at 1.7ghz with 1045mhz HBM2 and sticks to its BIOS temp values like a champ...

Doesn't seem like much of a clear cut difference b/n great and shitty ones to be honest, especially if you factor the prices in.
Posted on Reply
#39
jabbadap
[XC] Oj101If you look at the custom cards they have the SLI fingers at the top, some with a sticker over them saying that SLI is not supported. If you remove the heat sink you see that the board is identical in more ways than that.
Yeah that is a problem gtx1060s with gp104 chip, gtx1070 supports sli on the beginning so with these it does not matter.
Posted on Reply
#40
xkm1948
silentbogoI'm losing my hope for humanity. We get a decent bump in memory throughput, and people rant about it like it's a bad thing.
Some red guards would br happy if Nvidia gave 1070 slower VRAM and sell it higher.

You can’t convince brain dead fans.
Posted on Reply
#41
Vya Domus
silentbogoWe get a decent bump in memory throughput
Except we don't ? It's still 8Gbps chips just like previously and also just like in the case of the GDDR5X 1060. Make no mistake, no one would sell you a faster product with no increase in price.
xkm1948You can’t convince brain dead fans.
xkm1948Some red guards
Quality thread crapping and inflammatory comments from ya boi xkm1948.
Posted on Reply
#42
Darmok N Jalad
TurmaniaThis is getting ridiculous both Nvdia and AMD in GPU sector.I really hope Intel succeeds and be competitive in GPU sector or we are left with these two making mockery of us consumers.not that Intel is a saint but competition is good.
Right now, Intel hasn’t done much of anything, even with its own iGPUs, and those are tiny compared to high-end GPUs. What you can buy today does a great job at 1080p and 1440p, and even 4K is a possibility. If you want something like 8K, well, I hate to think how many transistors will be needed to game at such resolution, but however you look at it, that would get really expensive. I just don’t see Intel arriving to save the day anytime soon, and even if they did, they would charge Nvidia prices. AMD probably has a top-flight team of engineers in their GPU wing, and they don’t have an answer for Nvidia yet. It’s all uphill at Intel.
silentbogoI'm losing my hope for humanity. We get a decent bump in memory throughput, and people rant about it like it's a bad thing.
I’d like to see this memory on a RX 590. Reviews suggest it’s probably starving for bandwidth with its high GPU clocks buy only 256bit interface.
Posted on Reply
#43
Xazax
Just remember guys Cryptocurrency "Sales" only accounted for a few percent ;)

"NVIDIA's crypto revenues accounted for 7% and 3% of its second and third quarter revenues, respectively, so we can assume that percentage remains in the low single digits. Yet NVIDIA's total revenues rose 56% annually during the second quarter, 32% during the third quarter, and another 34% during the fourth quarter -- fueled by the strength of its gaming, professional visualization, and data center businesses. "
www.fool.com/investing/2018/04/17/nvidias-cryptocurrency-business-isnt-that-importan.aspx
Posted on Reply
#44
cdawall
where the hell are my stars
R0H1TSource for the 1080 boards being used in GTX 1060 or 1070 :confused:
Scroll down the news section for ten seconds and you'll find it.
Posted on Reply
#45
newtekie1
Semi-Retired Folder
Now I'm just waiting for the GTX1070 Ti w/ GDDR5X. That I'll definitely buy!
Posted on Reply
#46
Vayra86
xkm1948Some red guards would br happy if Nvidia gave 1070 slower VRAM and sell it higher.

You can’t convince brain dead fans.
The only thing red here is that -1 next to your comment. Are you collecting them lately?
Posted on Reply
#47
Raendor
ArbitraryAffectionHeya. Did you make an account just to say that? :)


There is an alternative. It's called the faster Vega 56 and 64. The former of which has sunk to, and below, MSRP here in the UK. Vega 56 can be had for £350 + free games. It's faster than 1070 in essentially everything with an exception to some horrifically NVIDIA favouring engines.
Lol, amd shill comes to advertise DOA space heater
I’d rather buy Pascal card again, if I needed one.
Posted on Reply
#48
medi01
RaendorDOA space heater
So, that's what Maxwell cards were, right, as that's where perf/w of Vega roughly is?

That delusional underdog hate is so pathetic.
Posted on Reply
#49
londiste
medi01So, that's what Maxwell cards were, right, as that's where perf/w of Vega roughly is?
While the DOA space heater might have been pure flamebait - Maxwell is from 2014 at 28nm while Vega is from last year and 14nm.
Posted on Reply
#50
megamanxtreme
Thinking about Space Heater, I would like a card that doesn't heat up my room when I game. Throttle it to 40C I say.
Posted on Reply
Add your own comment
Oct 28th, 2024 04:55 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts