Discussion in 'Frontpage news' started by Hilbert Hagedoorn, May 6, 2016.
So in your opinion amd designs their hardware to 'age like a fine wine' ?
More like they dunno wtf they're doing and it takes ages for them to figure out what they've made XD. Like making a car, but, from the start only 3 wheels spin. Later on, they figure-out how to make the 4th one spin aswell.
So the 1070 @ 379.00 Buck'A'Roo Dolla's is faster then the Titan X? I have my doubt's on this one but that's what nvidia cliams, Oh yeah? We shall see when Hilbert does the complete review on this one.
My eyes are on the HBM2 originally but that ain't going to be released for awhile now, Now my eyes are on the 1080, They should change the name though, lol, sounds like 1080 screen resolution, no disrespect but come on.
They should just eliminate the name Geforce, and just put Nvidia Pascal on the card instead, good for a change, been geforce for so long, but thats just me, Don't think it will ever happen though.
Really looking foward for theses cards, been waiting a long time, Volta seems interesting too, Speaking on topic with pascal, ain't the pascal supposed to be armed/equipped with a ARM cpu along with the gpu? Or is that the upper end model of the pascal?? I remember nvidia was putting some kind of cpu along with their new gpu in the future or did that all got canned?
Also gotta remember one more thing, the cards won't truly shine right off the get go, drivers needs to mature over time, probably talking maybe 6+ months tops to get this card really going, since its a brand new architecture.
Going with GDDR was a good gamble by Nvidia. Means they're on schedule with products available very soon. Only a short while back, almost everyone was doubting these cards would be coming so soon.
Business as usual at Nvidia.
Yeah same here, thought the cards probably would never be released this early, especially when the talks about the hbm2, had a feeling the cards wouldn't be equipped with them, well not at this time, probably end or beginning of next year perhaps, but not right now, not in mass amounts.
Was really looking into the hbm2, . got WAYS off before we see them, beside the p100 workstation card.
Dying to see reviews on the 1080 card and especially the 1070.
I can't wait to see gigabyte windforce version of the 1080, I'll take 2 or perhaps 3, never used tri before.
You have a point, but it's not all pink as you've said yourself. I mean at launch the 780Ti walked over the 290x and the roles were reversed on the meantime.
But do note that AMD has been on GCN since the 7000 series, this is the main reason their cards saw such good growth. Their drivers were bad to begin with, became better with time.
The point of this post is to not assume that AMD always makes hardware that lasts a long time. When they released GCN with the 7000 series they ditched Terascale really fast. Both AMD and Nvidia are known to drop their older architectures quite fast.
Maxwell was initially supposed to have the ARM CPU included, but plans changed, I don't remember the ARM CPU being a feature anymore in any of the upcoming architectures. This is what I remember from seeing NVidia roadmap slides.
Nah, that was just a rumor, it ended up being Tegra K1. They brought the GPU to the CPU and not the other way around. Nvidia never had any intention of putting an ARM CPU on one of it's desktop GPUs.
I'm pretty sure I saw roadmaps with ARM CPU included with Maxwell, they changed their plans for Maxwell very often, it was even supposed to have Unified Virtual Memory with Volta being the next architecture after that, but then they changed it to Pascal following Maxwell. They've just changed their plans multiple times.
They definitely changed Pascal/Volta, I agree with that. But I've never seen anything official from Nvidia saying that their GPU's would get an ARM processor on it.
Their ARM processor got a Kepler/Maxwell GPU on it though, with Tegra K1. I'm pretty sure what happened was that leaked info said that Nvidia was pairing a Maxwell/ARM processor together and rumor sites automatically assumed it was going on a desktop GPU. But it was actually just mobile.