Wednesday, May 18th 2016
NVIDIA GeForce GTX 1070 Clock Speeds Revealed
NVIDIA posted the product page of its upcoming GeForce GTX 1070 graphics card, confirming its clock-speeds, and related specifications. The card features a nominal GPU clock speed of 1506 MHz, with a maximum GPU Boost frequency of 1683 MHz. The memory is clocked at 2000 MHz (actual), or 8 GHz (GDDR5-effective), working out to a memory bandwidth of 256 GB/s. The company also rates the card's single-precision floating point performance at 6.45 TFLOP/s. Other key specs include 1,920 CUDA cores, 120 TMUs, and 64 ROPs. The GeForce GTX 1070 goes on sale, on the 10th of June, 2016.
123 Comments on NVIDIA GeForce GTX 1070 Clock Speeds Revealed
Medi01 simply stated that a GTX1080 is not that interesting to get compared to a GTX980Ti, thats just consumer advice (and perhaps more important regarding your remark, both are Nvidia cards....)
I think now it's safe to say next nvidia's card are once rumored gp104-150 as gtx1060ti and gp106 based gtx1060/gtx1050ti. I'm quite sure that nvidia will not let amd to take whole profitable low-mid range with their polaris gpus.
We all no doubt appreciate medi01's words of wisdom, and you're right they are both nVidia cards.... thanks for pointing that out.
Like I said, you stink of double standards.
*Can't believed I interrupted a game of DoD to respond to you're usual BS.
Second, your constant claiming that I am a fanboy is laughable... Guess I am the worst fanboy out there for the reds since I own a majority of products from anything but AMD. In fact, the only AMD product I own at the moment is the 3 R9 290X's in my main rig which currently has an Intel processor. Guessing my laptop (GTX 675m), backup/spare rig (i5 3570k 550ti), server (Intel Xeon 5670x2, Nvidia GTX 950), that all have the opposite sides products must really make me a total fanboy. Getting kinda tiresome that by your standards anyone not insulting AMD on a constant basis is a fanboy of the red team.
Hell just bought my stepfather a cheap HP 255 laptop, fully AMD powered.... yeah your boring me too. You seem to single me out these days, I guess it's gotten personal now? I'm flattered and all that but I'd prefer if you put me on ignore and do us both a favour.
Anyway, looking forward to see what custom AIB GTX 1070's can do. :cool:
Hope there will be a smaller version of the card, i don't feel like slapping a huge card on my matx rig, though i hear some not so good feedbacks on the prolonged 970 minis of Gigabyte and Asus
One big concern I still have for the GTX 1070 is the power limit. Wonder how far it will go?
970 and 980 were much more closely clocked at stock and therefore had much more similar binning process.
Statistically 1070 should clock worse than 1080, but highest clocking pascal should be cherry picked 1070 on account of 5 disabled TPCs.
edit: I just realized you were talking about the power limit
Promises look quite optimistic (I was worried they kept it mum for a while). With "we'll close the gap" and "we never been so close".
(although I can remember times when they were ahead, it was more than 10 years ago so probably doesn't count)
Anyhow, if AMD delivers, you can get your beloved Intel/nVidia chips for less, why not cross finger for AMD for that alone?
I, for one, will likely buy 480(x) (to replace 380) and later on likely Zen, (to replace i5 750). Well, on a driver front, especially when talking about anomalies that, for instance, 290x keeps becoming faster vs 780Ti for the third consecutive year, people normally accounted that for AMD needing more time to develop good drivers. So theory goes that nVidia drivers are good and fast right away. That graph was 1080 vs cheaper previous gen card with model number that ended with 80, not vs Tx.
Curious part of it is, graphs for 1080 vs 980 are exactly the same as for 1070 vs 970... =))) Isn't it nearly always the case with overclocking?
There is certain threshold when upping this and that reasonably increases power consumption, but at some point power consumption grows too quickly.
Now, 980Ti was a brilliant OC-er and hands down best card high end tier of the last gen. (yep, I said that. Actually, I have no problems saying that, it's just some nZilla fanboi go defensive for no reason when reading my comments). It could easily get +40% of the stock, and even 50% wasn't that rare.
With 12% OC 1080 simply doesn't cut it. It might in the future, but not that there isn't much time left if Vega is coming in Oct, as rumored.
Oh, that if GloFo / Samsung 14nm doesn't fruck up, which, from what I've heard, namely:
a) Apple's TSMC 16nm chips are superior to the same chip on Samsungs 14nm (I, frankly, don't get how it could be "the same chip" given the differences in the process, but, oh well)
b) Nobody really forced nVidia to go with TSMC. They would NOT risk their high end dominance to simply get cheaper chips, so at the very least, there is no clear superiority in what Samsung's 14nm is doing
c) Some vague badmouthing about GloFo underperforming and underperforming
it could well do. There is no Dell XPS with AMD APU/CPU.
Also no high end Lenovo either.
HP might have some, as they (Compaq) tend to favor AMD.
Did i said to anyone who is going to buy a new vga now go and take 980ti and not 1080?????????
My point is if you already have a custom 980ti now, there is no need to update, 30% is only when compared to stock 980ti (1ghz) the cards i posted above ARE of course oced BUT factory and working like that STOCK, both can oced at least 1.4ghz, Or users have to underclocked their cards because reviewers wnat to show what it have to be shown
R u mr genious or mister dumb?
I'd like to know!