Dictator93
Member
Isn't Fiji largest die they have ever done?
Doesn't the R600 stil have that crown?
Isn't Fiji largest die they have ever done?
If we believe rumors about Fiji availability then it's basically same as not having it release at all during 2015. 30k cards (over whole 2015) is absolutely nothing and basically same as not having stock at all, which just increases AMD's bad rep and will have negative impact when moving into 2016. Is there any economical sense in "Well we released world's first GPU with HBM, glhf with finding any store on this planet that has one in sale after first 3 days for rest of the year" release? Backlog on orders would be mind blowingly big and people just would pick 980Ti instead.
In such low yield scenario it would make more sense, to me at least, do light refresh on 200 -series for rest of 2015 and do 14nm Fiji release in 2016 with higher yielding HBM 2.0. Basically what NV is aiming to do with Pascal.
1. I hope it has those things. Buuut... we shall see. That would mean it has to be tonga, right?I'm skipping this generation waiting for next year so my dog ain't really in this fight but I do have some comments about this.
1. If it is a straight rebrand with more ram, to me that does suck, I just see no reason why they wouldn't add the tessellation improvements and color compression in these.
2. What does it say about Nvidia that AMD's old hardware still competes with their new hardware, relatively.
2. What does it say about Nvidia that AMD's old hardware still competes with their new hardware, relatively.
Doesn't the R600 stil have that crown?
That they are still using 28nm
AFAIK, yes. They probably felt that they had no choice this time around.
I'm skipping this generation waiting for next year so my dog ain't really in this fight but I do have some comments about this.
1. If it is a straight rebrand with more ram, to me that does suck, I just see no reason why they wouldn't add the tessellation improvements and color compression in these.
2. What does it say about Nvidia that AMD's old hardware still competes with their new hardware, relatively.
What's the rumored pricing? I'm in the market for a new card here soon, so I'd like to know what I've got coming for me.
Where did you hear that?What I heard they went big because it had no effect on performance to direction or another, but made chip cooling just that much easier and better.
Tonga is pretty good, everything before it, not so much (workable at low tesselation factors).1. Doesn't 200 -series handle 16x tessellation fine? Is anything gained in IQ when going above that? Just look at HairWorks with its default 64x vs. custom 16x
Honestly it is quite amazing how far the GCN1.x microarchitecture has carried them. In the same timeframe Nvidia has had two revisions of Kepler and two revisions of Maxwell. They have always been the "less good" but it has remained competitive.2. That AMD HW is actually rather decent stuff with good lifespan?
They said 30K for the Fury X. There's still the Fury Pro, a cut down part.
If the Fury X is a halo part priced at $800+ I don't see AMD really needing to have a ton of units. Fury Pro will drive most of the sales if it's significantly cheaper and offers most of the performance, akin to the 980 Ti versus Titan X.
Fury Pro would be the more expensive card, no? What with it being a "pro" card and all. Unless AMD is gonna start labeling gaming cards with a "pro" moniker. For pro-gamers I suppose.
Fury Pro would be the more expensive card, no? What with it being a "pro" card and all. Unless AMD is gonna start labeling gaming cards with a "pro" moniker. For pro-gamers I suppose.
Nah. It is related to AMDs gpu codename scheme, where Fiji Pro is the cut down version of Fiji XT. We only suspect that the cards will be named Fury and Fury X.
What I heard they went big because it had no effect on performance to direction or another, but made chip cooling just that much easier and better.
1. Doesn't 200 -series handle 16x tessellation fine? Is anything gained in IQ when going above that? Just look at HairWorks with its default 64x vs. custom 16x
2. That AMD HW is actually rather decent stuff with good lifespan?
I don't think anyone is suprised if FuryXT is priced aggressively against 980Ti and FuryPro then step below that.
You'd only be disappointed if you were not following the leaks/rumors. The 300 series being rebrand has been known since I dont know, maybe March?
It's the largest die ever from AMD/ATI.Isn't Fiji largest die they have ever done?
That joke journalist site - wccf. Bigger die equals easier to cool!!!!!1Where did you hear that?
I was not aware that AMD was a person whose feelings I was hurting. I thought I was just offering criticism of things a corporation called AMD have done or are doing, I didn't know AMD was like someone's best friend.
You haven't offered analysis or arguments to support your position. It doesn't matter what AMD has done, you interpret it as some kind of failure without any basis. There's a word for that.
I've provided more analysis than most people in this thread have. My interpretations have always been my own and I never pretended anything I said was more than my own opinion. But you seem to have a problem with anybody who criticizes your favorite video card company. There's a word for that too.
You guys need to stop getting so personal and chill.
Anyways, whens event happening? Wasn't it on the 15th?
Your analysis has always consisted of stating what AMD is doing and insinuating it is the result of incompetence. You never back anything up. You assume that which you claim to be proving. I'm calling you out because you are a consistent bad actor in this discussion. I recognize that your mental state may prevent you from ever recognizing the shape of your behavior, but you don't get to just pretend I'm a wounded fanboy and you're above it all. Anyone who has followed your posts can see what prejudicial waste of space they are. You'll notice I'm only talking to you. There are plenty of other nVidia fans in this thread looking sideways at the AMD news and rumors. You're the only one here making an ass of yourself.
Never seen a post try this hard to insult someone and at the same time skirt the ban rules. And over videocards?
I have a question. Since the 3XX cards are seemingly straight up rebrands, stock should be available immediately and in abundance right? A friend's 7970 died yesterday and we were discussing what his next options are. Now we live in a country where new hardware usually takes a few weeks to show up and he can't go that long without a new card so we settled on a 970. But if vendors are able to get stock into stores quickly then maybe a 390 would be on the table.
Your analysis has always consisted of stating what AMD is doing and insinuating it is the result of incompetence. You never back anything up. You assume that which you claim to be proving. I'm calling you out because you are a consistent bad actor in this discussion. I recognize that your mental state may prevent you from ever recognizing the shape of your behavior, but you don't get to just pretend I'm a wounded fanboy and you're above it all. Anyone who has followed your posts can see what prejudicial waste of space they are. You'll notice I'm only talking to you. There are plenty of other nVidia fans in this thread looking sideways at the AMD news and rumors. You're the only one here making an ass of yourself.
Yes, please stop feeding him (and that includes you Brad).
Looks like the new cards are more power efficient - http://forums.anandtech.com/showpost.php?p=37481186&postcount=1456
Plus Dave at B3D is hinting to not read into (TDP) stickers on boxes.
I think some of Wccftech's articles have pointed towards a lower TDP for at least the R7 360 too.
They're doing exactly what Dave is asking not to do - read the listed TDP on the box.I think some of Wccftech's articles have pointed towards a lower TDP for at least the R7 360 too.
Check my link, comparison is between 390 and 290, both are same CU count.Seems that the 390 isn't 100% a Hawaii chip if energy efficiency improved a lot.
https://www.reddit.com/r/Amd/comments/39u751/xfx_r9_390_temps_in_furmark/
They're doing exactly what Dave is asking not to do - read the listed TDP on the box.
Check my link, comparison is between 390 and 290, both are same CU count.
Yes, please stop feeding him (and that includes you Brad).
Looks like the new cards are more power efficient - http://forums.anandtech.com/showpost.php?p=37481186&postcount=1456
Plus Dave at B3D is hinting to not read into (TDP) stickers on boxes.
Yes, this makes sense and anyone following GPUs should know this. Apparently pointing this out is "desperate".Is this about the 375W figure that showed up on some boxes? It seemed to me that was just the maximum power 2x8-pins could deliver instead of the actual TDP.
The drop seems pretty significant though. I wonder if it's due to better binning or they went for a respin.
Is this about the 375W figure that showed up on some boxes? It seemed to me that was just the maximum power 2x8-pins could deliver instead of the actual TDP.
The drop seems pretty significant though. I wonder if it's due to better binning or they went for a respin.
If they're using GF instead of TSMC, it would make sense that the gpu would have different characteristics even if the IC design remained the same.
No doubt need to wait for more concrete info...
But, so far the only remotely interesting models are Fiji XT/Pro (obviously) & the as-yet unannounced possibility of a 380X (full Tonga/Antigua). A 380X 4GB w/ 2048 shaders in the ~ $250 - $275 price range would be an instant purchase for me. Every other model below Fiji is a rebrand. And if that turns out to be the case the 300 product stack as it stands is a total yawner. Sure, it'll be nice if there are some efficiency gains at each level in the stack. But that alone isn't going to be enough to recover market share. Given the 380X is easily AMD's best chance to compete at or near the all-important ~$250 mid-range price-point I don't see a good reason for holding the uncut Tonga back. Surely Apple can't have some sort of actual timed exclusive agreement on this model can they? I keep hearing this as one possible reason why the 380X hasn't been announced yet. But that'd be just about the dumbest move ever for AMD if that is actually the case. There is a gaping price hole/performance vulnerability between the GTX 960 & GTX 970. It's about as obvious as you can get. AMD could attack there & clean up for 6 months until nVidia can roll out a GTX 960Ti to compete. Fiji is going to be amazing. But, you don't win market share with $750 videocards at the top & a bunch of 3rd gen rebrands at every other level in the stack. The top end card is just bragging rights/e-peen pricing territory. You win significant market share with your $250 and under parts. So the fact that they're holding their best potential model back is kinda mind-boggling....
AMD PR damage controlling pretty hard calling the rebrand stuff silly rumors. Fury better be awesome because the 300 series seems pretty lame.
AMD PR damage controlling pretty hard calling the rebrand stuff silly rumors. Fury better be awesome because the 300 series seems pretty lame.
Wait till launch day - I can't wait for the storm of juniors posting about great gpus with 8gb of vram vs 3,5GB 970![]()
AMD PR damage controlling pretty hard calling the rebrand stuff silly rumors. Fury better be awesome because the 300 series seems pretty lame.