Possibly why their ultrasharp lines are so good, kinda like Apple being late to the game on a lot of featuresAll I want is Dell to introduce Freesync in their Ultrasharp lineup. Once that happens I'm getting their updated U25-whatever 25" 1440p 60Hz. Though knowing Dell, it's going to be a long wait.
So roughly 10% slower than a GTX1080TI Founders Edition??
Heh, how can you know this?
If you're referring to Mockingbird's claim that Vega won't have much overhead, I'd have to agree given what we know about GLOFO and AMD's propensity over the past several years to push CPUs and GPUs to the max from the factory leaving little overhead to OC.
To me what's still up in the air is out of the box performance. I'm hoping for the best but expecting the worst but who knows, maybe AMD will pull off some magic like they did with Ryzen.
I've never really seen this as an issue.. Or understood it really.
Why is it seen as a bad thing that Radeon extracts almost the maximum performance out of their graphics cards, and on the flipside why is it okay for Nvidia to underachieve on their stock clocks?
(obviously if Vega and 1080ti ended up same speed at stock but 1080ti could OC higher I guess there's an argument, but even then, why doesn't the 1080ti just come higher clocked)
No.There's no fundamental difference in neither the content nor the form of anything in your quote. It's all speculation (some backed up by various rumors, some not), but also all presented as fact. Look at how it's written, the language used:
(my emphasis)
Those are not word forms commonly associated with expressing neither opinion nor speculation, unless prefaced with something clarifying that what follows is just that. I don't see anything like that anywhere here. Those are word forms used for expressing facts, things that are, not things that might be. Which is the entire root of this discussion. None here are opposed to neither rumors nor speculation (that's what this thread is for, after all), but expressing it as if it were indisputably true - and refusing to provide sources to boot! - is just silly. I mean, adding "I think" or "it sounds reasonable that" into a sentence here and there would make this whole problem go away. But that's the job of the writer, not the reader. And failing to do so reflects a rather disturbing relationship with knowledge, facts and truth on the part of the writer. #alternativefacts
I'm glad someone has decided to take on the role of w3rd and bring back the "making sh*t up and proclaiming it as fact" in this thread. It's been too long. Sources, you say? Bah, humbug! Mockingbird is the absolute authority on anything and everything AMD. How dare we question his indisputable authority by asking for 'sources'? Ludicrous!
if vega was just a bigger polaris it would have been released last year. so no you cant just simply take polaris and guess its size or perf.
I've never really seen this as an issue.. Or understood it really.
Why is it seen as a bad thing that Radeon extracts almost the maximum performance out of their graphics cards, and on the flipside why is it okay for Nvidia to underachieve on their stock clocks?
(obviously if Vega and 1080ti ended up same speed at stock but 1080ti could OC higher I guess there's an argument, but even then, why doesn't the 1080ti just come higher clocked)
For most consumers it doesn't matter and they don't care. For me, it's a useful metric since I OC and mine and care about power consumption. Power consumption for GPUs is a biggie. One of the reasons NVIDIA's GPUs are so much more efficient out of the box is that they are typically clocked at a somewhat efficient level. Look at AMD with the 480, pushed up against the process's ceiling and consuming far more juice and creating far more heat than it would have if clocked at a more reasonable level.
Those of us who mine ETH know how efficient the 480 can be when underclocked and undervolted and while not as efficient as NVIDIA, they do pretty good.
My 1070 sustains ~ 2050mhz which is really quite amazing when you think about it. For those of us that like to extract every bit of power out of a GPU, NVIDIA's cards allow that, not so much with AMD anymore although I loved what the 7000 series could do and still have a 7970 Toxic laying around that served me very well.
If you're referring to Mockingbird's claim that Vega won't have much overhead, I'd have to agree given what we know about GLOFO and AMD's propensity over the past several years to push CPUs and GPUs to the max from the factory leaving little overhead to OC.
To me what's still up in the air is out of the box performance. I'm hoping for the best but expecting the worst but who knows, maybe AMD will pull off some magic like they did with Ryzen.
As an example, this is what we're looking for:
this is not:
AT Modertor ElFenix
I've never really seen this as an issue.. Or understood it really.
Why is it seen as a bad thing that Radeon extracts almost the maximum performance out of their graphics cards, and on the flipside why is it okay for Nvidia to underachieve on their stock clocks?
Doesn't sound like a miracle to me, just a good but expected job. I doubt AMD designed it for any less than it launched at, and the GloFo process had time to mature.AMD really did pull off a small miracle considering that last year AMD was having trouble getting Ryzen to operate over 3 GHz
Actually, Intel clocks a full GHz higher than Ryzen, so I don't see what you mean.The FUD about Glo Fo's 14nm process being poor is not going to stick. It's easy to prove as FUD just by pointing out that Ryzen is on par or better than intel's comparative chips on their latest 14nm node.
We already know how efficiently the transistors switch at especially 3.5GHz or less, with it losing some efficiency towards 4GHz. So the process is certainly capable of operating at higher frequencies than what AMD's GPUs currently operate at, however the prior architectures are not designed to run at frequencies higher than what Polaris is running at.
Higher frequencies are part of Vega's design goals. If it was just Polaris running at the same frequency as the Pascal competitor, well there's a 30% performance advantage right there.
Maybe we'll know more at AMD's financial analyst day in a week.
The FUD about Glo Fo's 14nm process being poor is not going to stick. It's easy to prove as FUD just by pointing out that Ryzen is on par or better than intel's comparative chips on their latest 14nm node.
Doesn't sound like a miracle to me, just a good but expected job. I doubt AMD designed it for any less than it launched at, and the GloFo process had time to mature.
Actually, Intel clocks a full GHz higher than Ryzen, so I don't see what you mean.
As far 14nm vs 16nm, we can look at 1050 Ti. The thing maxes out at around 1.9GHz, as compared to the 2.1GHz of its 16nm brethren.
No, it is a small miracle when you considered what limitation AMD has to deal with.
Yes, I am talking about GlobalFoundaries.
Signing the Wafer Supply Agreement is one most shortsighted thing AMD has ever done.
Its funny. Scott Wasson has in-depth explanation on reasons why Nvidia hardware clocks so high.Doesn't sound like a miracle to me, just a good but expected job. I doubt AMD designed it for any less than it launched at, and the GloFo process had time to mature.
Actually, Intel clocks a full GHz higher than Ryzen, so I don't see what you mean.
As far 14nm vs 16nm, we can look at 1050 Ti. The thing maxes out at around 1.9GHz, as compared to the 2.1GHz of its 16nm brethren.