What if it's a sale to a youtuber who does a critical video about Vega, with the conceit being that it's more useful as scrap than as a video card?
What if it's a sale to a youtuber who does a critical video about Vega, with the conceit being that it's more useful as scrap than as a video card?If you truly believe that a sale in which I trash the GPU is equivalent to a sale in which I use the GPU, potentially review the GPU, etc. etc. then there really isn't a point in me bothering to explain it. It's just a hopeless endeavor since the basic principles are beyond the grasp of the person attempting to understand.
That's why financial conversation sucks to have on here, because something this basic is apparently hard to grasp. It's not an opinion, it's a fact.
If you think a sale to JayZTwoCents is equivalent to a sale in which I buy the GPU and dump it in the trash. I can't even begin to have an intellectual conversation with you on this matter.
Lets just move on to tech stuff.
What if it's a sale to a youtuber who does a critical video about Vega, with the conceit being that it's more useful as scrap than as a video card?
My use of " conditions" was appropriate, "a restricting or modifying factor" as defined by Websters. Your later assertion miners don't provide the same word of mouth advertising as gamers was not implicit in your original statement, largely because it is unsupported by anything but opinion. Indeed, evidence is to the contrary as AMD parts have become wildly popular in the mining community in the absence of widespread reporting on their superiority for this function.No, I'm not adding other conditions. Conditions isn't even the right term, which again is why this is pointless and is the last post I'm making on the topic because we're simply not having an equivalent conversation. This is a similar conversation to a person who claims the world is flat because that's all they can see. Without being able to understand how to make that next mental leap to understand anything beyond the initial thing you first see, there really is no hope for further education on this matter.
Wasn't the original launch for this supposed to be Q4 last year?
Why is it difficult to use eg two gpu on an interconnect like infinity fabric?
I mean we run highly branced code like games on 8c cpu with great results. Spanning dies and sockest.
Of course they lost out. The problem was that AMD have been selling the same card for far too long, so why buy a new 390 when you can a used 290. Still the same now with the 580 and 480.Its impossible for AMD to have lost out on sales if people were buying them used, as that means AMD had already sold the card.
What hurts them is the cards price skyrocketing while the retailers are getting the money and not AMD charging more. So the price is higher than they want and thus worse on price/perf while they don't benefit from the extra income (retailers do)
Vega x2 was leaked before, it exists.It is about connecting the dots.
Caspian & Cream event, Bethesda said they have near 100% scalability. AMD engineers already confirmed this, and not only that, the ability to de-couple the gpu's and use them for different tasks, etc. (Real time TruAudio and other 3D environmentals).
Not sure why so many people are feigning ignorance on any of this. This is the future and something AMD has been aiming at since acquiring ATi, plus they already have their APU as proof of design, as to what they can do with Vega on fabric (or refined & on Navi).
Non of that matter's to the joe consumer really. But upon their release joe will see Radeon Vega as premium choice. Which will also elevate the Radeon brand, giving it massive headspace.
Might as well get on board (wrap your head around) that AMD's "secret sauce" is not hype, it is a revolution taking place. And We (as consumers) are about 3 months away from watching the floodgates open. Some just can't see it.
Obviously single Vega (& cut downs) are "entry level" to the 4k Gaming Market.
And being the cheapest tier the initial premium for Vega's HBM2 will be costly step for many consumers to make. But also worth it, because of the "other aspects" of what Vega's platform brings. One of those aspects is the simple fact that HBM2 will not be about high frames, but ultra maintained minimums. Another aspect is FreeSync2 is available on Vega's platform.
Play around with and pretend knowing for sake of argument:
- baby vega is single die gpu
- big vega is double die gpu
Ponder on that^ strategy for a bit....
Then envision Vega's entire product stack/sku and what it is going to look like in 6 months from now. ($399 ~ $1,200?)
tick
tock
Holiday shopping Season!
"and you get fabric.. and you get some fabric.. everyone gets fabric..!"
Economics would suggest, that by December not so right "cut down" x2 version of Vega to be the best buys. But still way more capable than a $800 product from a competitor.
Can you wait..? Is all...!
Vega x2 was leaked before, it exists.
What we don't believe is that they're connected on an interposer and with infinity fabric, and can act as a single GPU.
That's a leap right there.
And Radeon Pro Duo is what? Not Polaris 10 X2?I want to also point out here that even Polaris X2 has been leaked, but we are (except we aren't, actually) still waiting for it. I mean, it is possibility I have no intention to deny. What I'm denying here is that "leaks" make Vega x2 a certainty :>
Excuse me? AMD showed a product roadmap at their capsaicin event last March that showed Vega launching December 2016. (And I bet last summer had a few "wait for Vega" posts that never would have happened had we known Vega was a year away)Your post make sense. I quote just this because Vega has always been a 2017 product in AMD roadmaps
Excuse me? AMD showed a product roadmap at their capsaicin event last March that showed Vega launching December 2016. (And I bet last summer had a few "wait for Vega" posts that never would have happened had we known Vega was a year away)
And expecting it to be launched for consumers is another leap, considering the past two dual GPU cards from AMD have been pro/datacenter only.Vega x2 was leaked before, it exists.
What we don't believe is that they're connected on an interposer and with infinity fabric, and can act as a single GPU.
That's a leap right there.
LOL, type "Capsaicin GPU roadmap" into Google and take your pick from all the major tech websites posting that roadmap. (Including Anandtech)Link?
LOL, type "Capsaicin GPU roadmap" into Google and take your pick from all the major tech websites posting that roadmap. (Including Anandtech)
You may "want" to change history, but back then you were probably posting "wait for Vegas.
I think he's referring to the one on the bottom, which by many was (obviously in hindsight mis)read as "Vega before 2017" due to the "Vega" box being slightly to the left of the "2017" text. Considering that roadmap is vaguely labeled at best, it's important to point this out. Not to mention the "projected roadmap, subject to change" caveat, that may or may not be relevant (depending on whether Vega is actually delayed compared to the original launch schedule or not).
You have to try harder.
You can surely provide a link to that statements made by me as you provided the links proving your fake news.
Yes. Even the Anandtech article notes the left edge of the Vega box falls in 2016 and their surprise it may be this early after Polaris.I think he's referring to the one on the bottom, which by many was (obviously in hindsight mis)read as "Vega before 2017" due to the "Vega" box being slightly to the left of the "2017" text. Considering that roadmap is vaguely labeled at best, it's important to point this out. Not to mention the "projected roadmap, subject to change" caveat, that may or may not be relevant (depending on whether Vega is actually delayed compared to the original launch schedule or not).
I think he's referring to the one on the bottom, which by many was (obviously in hindsight mis)read as "Vega before 2017" due to the "Vega" box being slightly to the left of the "2017" text. Considering that roadmap is vaguely labeled at best, it's important to point this out. Not to mention the "projected roadmap, subject to change" caveat, that may or may not be relevant (depending on whether Vega is actually delayed compared to the original launch schedule or not).
They might have, but that slide is clearly formatted vaguely on purpose. After all, concrete roadmaps look more like this:Yes. Even the Anandtech article notes the left edge of the Vega box falls in 2016 and their surprise it may be this early after Polaris.
Vega x2 was leaked before, it exists.
What we don't believe is that they're connected on an interposer and with infinity fabric, and can act as a single GPU.
That's a leap right there.
Can you explain what general efficiencies you think you would get with a dual GPU card that would cause it to use less power than two individual GPUs?I was gonna give you the benefit of the doubt and assume you've let w3rd to steer you off the course of reality, but then you reply to me with this biased nonsense. I don't think I should add anything, it would just make it personal, instead I just rest my case
Sent from my VTR-L09 using Tapatalk
..... No need to write a book for this queation, please. The voltage/wattage curve is simply not linear. The efficiency does not come from it being dual GPU itself, but rather how they always design them. And please also don't try to convince me with the outliers. It's like saying NVIDIA is a shit company just because of the Titan Z.Can you explain what general efficiencies you think you would get with a dual GPU card that would cause it to use less power than two individual GPUs?
In applications where performance scales perfectly (IE, crypto mining) I've found that a dual GPU card uses pretty much twice the power of a single card. For example, my water cooled Devil13 dual 390 card used pretty much the exact same amount of power my dual WC 390s use. The exception to that rule that I've seen is the 5970 which used less power for the same hashrate vs two 5870s, but that was probably due to the extreme binning ATI used on those dies.
In applications where you don't get perfect scaling (IE, crossfire gaming) your power use for a dual die card might be less than twice a single die card, but that has little to do with it being a dual GPU card; you'd get the same lower power draw with two single die cards in crossfire.