AMD: Pirate Islands (R* 3** series) Speculation/Rumor Thread

Maybe some restrictions regarding games anda settings that use more that 4 GB of memory.

Not that I'm agreeing that this happens but if it did, then restricting benchmarks to 4K would push things very much in favour of AMD while 1080p would favour Nvidia. This does account for some of the huge variations we see in performance between Nvidia and AMD. That and the games and settings used. I do wonder at the logic of these selections from some sites. Often a mixture of ancient or fairly obscure games at weird settings. Personally, for high end video cards I'd test everything maxed out (minus Nvidia exclusive Gameworks features or features which have an extreme bias towards and are sponsored by one particular vendor (hair works)) apart from MSAA at 1080p, 1440p and 4K in all of the games below and include a performance summary/average across all the games at each resolution. That's about as unbiased as you can get IMO. These games have been selected purely on the graphical requirements and popularity while trying to get a mix of genres in there.

GTAV
Watch Dogs
AC: Unity
MGS5: Ground Zero's
The Evil Within
Ryse
Project Cars
The Witcher 3
Dragon Age Inquisition
Farcry 4
Crysis 3
Wolfenstien Old Blood
Battlefield 4/Hardline
Call of Duty Advanced Warfare
Metro Redux
MGS5: Ground Zero's
Middle Earth: SOM
 
Exchange between Ryan and James Prior of AMD ;)

https://twitter.com/cavemanjim/status/609524426887946240

Sorry for being offtopic but it was vaguely related to this launch and earlier discussion.
Yay for mud wrestling game on the Internet! Bring out the popcorn!

But I'm just surprised that a PR expert (or so he says on his own LinkedIn page) would do it just days before one of the biggest launches in years. Like truthfully answering a question about not having DX 12_1, it doesn't sound like the thing you'd do to stay on message...
 
What makes this one of the biggest launches in years? Just curious.
Fiji is a bit of a make or break for AMD's GPU division? I don't remember the last time they made a really big deal about a GPU launch: pre-announcements, technical preview of technology parts etc.
Maybe you don't think it's a big deal, but it seems to be one for them... And the last time they had a major GPU launch was the 7970. That's over 3 years ago.
 
They did pretty much the same kind of press and marketing with Hawaii's launch but they didn't have the technical preview that they have now with HBM. I don't think this is any bigger than Hawaii, its the same thing with a new chip at the top and everything else being shuffled down. I assume next year will be a whole new line and that's when the big launch is.
 
So Ryan is annoyed that AMD is doing a publicity event without him?
Maybe he's annoyed AMD is doing a publicity event with fanboys instead of actual tech press (such as the state of that is on the internet)? I must say, if that's actually what AMD's done, it's incredibly disappointing, approaching pathetic really.

But I'm just surprised that a PR expert (or so he says on his own LinkedIn page) would do it just days before one of the biggest launches in years.
He's sure being very passive-aggressive, borderline overtly hostile in his replies. It's not good, for sure.
 
They did pretty much the same kind of press and marketing with Hawaii's launch but they didn't have the technical preview that they have now with HBM.
Matter of perception, I guess. But since Hawaii was launched 21 months ago, it could (barely?) qualify for 'in years'. Or also, I said "one of the biggest...", with Hawaii the other one. Either way, I win! ;)
 
These cards should be in reviewers hands considering that they are sitting in the warehouses of most sellers now. Come tuesday, these cards are launching and at best reviewers might get the cards on that day. Considering some reviewers don't have tracking numbers even, this is disrespectful in the sense that how can AMD expect quality reviews if people are not allowed atleast a week to play with the cards, particularly since it's 4-5 cards being launched at once.

The worst is if reviewers are only given fiji cards to create hoopla for the new series. AMD sort of did this already when they did an NDA controlled review where reviewers for their APU's where only allowed to released benchmarks for the graphical portion vs intel chips and not the CPU performance.

Giving these cards to biased fanboys who will undoubtly give positive press about the cards without properly evaluating the cards is a slap in the face about the cards is a slap to any objective reviewer who does this for a living.
 
From a guy that bought 390s from bestbuy:

9Fwd2nz.jpg


https://www.reddit.com/r/Amd/comments/39qol9/r9_390_gpuz_for_those_that_asked_yesterday/

And then another guy who bought a 390X (from bestbuy as well) benching it in 3dmark:


He also has an unboxing video however the point is that the 390X is seen as a 290X by futuremark's sysinfo.

I'm not sure if there were people here still doubtful of the hawaii rebrand theory but anyways, the "grenada is gcn 1.2 on glofo" talk that I've seen elsewhere such as anand should probably be debunked enough by now.

Also I would definitely understand if AMD decided not to sample the non Fiji 300 series cards to reviewers. They want to keep away from the rebrand talk if possible. Reviewers are just going to call them out on it and not sampling them means that they can cash in on the immense 300 series hype that's going on at the moment on reddit etc. A lot of people who don't follow these things so closely still seem to think that the 390X and 390 are Fiji cards and that the ~$400 prices are outlandishly good because of it. Why kill the hype with downer reviewers?
 
If 300 series are all rebrands, not giving them to reviewers makes perfect sense. Give them the halo Fury X create the hype and quietly release the 300 series riding on the hype wave.

Isn't that what halo products are for?
 
If 300 series are all rebrands, not giving them to reviewers makes perfect sense. Give them the halo Fury X create the hype and quietly release the 300 series riding on the hype wave.
Isn't that what halo products are for?

Actually they'd probably be better off giving them to reviewers so informed decisions prior to purchase can be made. Perhaps AMD prefers "product returns" over "no sales", but in the end it's all the same as the returns have started.

..my local best buy has the 390x in stock ..$449.00. picked one up and ran some benchmarks. a little disappointed it doesn't beat my reference gtx 980. Oh well , my fun is done and now im boxing it up and returning it because it has all the same isssues my 290x does in witcher, project cars and dirt rally. -what an uber let down on performance
http://forums.guru3d.com/showpost.php?p=5095851&postcount=51
...
I'm actually at the local mall sitting here eating but when I get home I'll post some quick in game benchmarks. I know drivers will mature over time but I wasn't impressed with the performance of the 390x. it barely beat my 290x in fps in witcher , project cars. in those games it was also slower than my 970 & 980 by a good margin. Dirt rally still has the dotted transparency bug with the 390x.. so does GTA V .- I sure hope the fury has better driver results because these 15.5's are complete crap with the rebranded 3 series.
http://forums.guru3d.com/showpost.php?p=5095858&postcount=53
.
 
Not sure what is going on there. OpenCL 2.0 compliance occurs with Bonaire and later.
Well. it's „your“ hardware and „your“ drivers. So you should know what's going on there, shouldn't you? Why would the drivers report this feature if it isn't there?
 
First Euro-dealer prices and delivery dates for Radeon R300 models from Gigabyte, MSI and Sapphire

Gigabyte (delivery on 18 June)

Gigabyte R9 390X 8GB Gaming G1 € 499.90
Gigabyte R9 390 Gaming G1 8GB € 419.90
Gigabyte R9 380 Gaming G1 4GB € 289.90
Gigabyte R9 380 WindForce 2x OC 2GB € 259.90
Gigabyte R7 370 WindForce 2x OC 2GB € 179.90


Sapphire (delivery date: June 16)

Sapphire R9 390X Tri-X OC 8GB € 549.90
Sapphire R9 390 Nitro 4GB € 549.90
Sapphire R9 380 Nitro Dual-X OC 4GB € 299.90
Sapphire R9 380 ITX Compact OC 2GB € 269.90
Sapphire R9 380 Dual-X OC 2GB € 269.90
Sapphire R7 370 Nitro Dual-X OC 4GB € 214.39
Sapphire R7 370 Dual-X OC 2GB € 189.90


MSI (delivery date: June 16)

MSI R9 390X Gaming 8G € 549.90
MSI R9 390 Gaming 8G € 459.90
MSI R9 380 Gaming 4G € 299.90
MSI R9 380 Gaming 2G € 269.39
MSI R9 380 2GD5T OC € 249.39
MSI Gaming R7 370 4G € 219.90
MSI Gaming R7 370 2G € 189.90
MSI R7 370 2GD5T OC € 184.39
MSI R7 360 2GD5 OC € 139.39

http://www.3dcenter.org/news/erste-...adeon-r300-modellen-von-gigabyte-msi-sapphire
 
They did pretty much the same kind of press and marketing with Hawaii's launch but they didn't have the technical preview that they have now with HBM. I don't think this is any bigger than Hawaii, its the same thing with a new chip at the top and everything else being shuffled down. I assume next year will be a whole new line and that's when the big launch is.

The fact that it's the launch platform for the biggest revolution in graphics memory since GDDR1 makes it one of the "biggest launched in years".
 
Actually they'd probably be better off giving them to reviewers so informed decisions prior to purchase can be made. Perhaps AMD prefers "product returns" over "no sales", but in the end it's all the same as the returns have started.

.

Some may return it. Majority probably wont. Anything AMD moves is better than having a huge stockpile of 200 series dies ;)
 
The fact that it's the launch platform for the biggest revolution in graphics memory since GDDR1 makes it one of the "biggest launched in years".

Little in between, optimal would been HBm2 and a die shrink obviously.
Fury advanced HBm tech is highly exciting news.
 
Well. it's „your“ hardware and „your“ drivers. So you should know what's going on there, shouldn't you? Why would the drivers report this feature if it isn't there?
Mine, personally? No.

The vagaries of OpenCL conformance vs what a driver may expose or what an application may say is reporting is a detailed thing. What I care about is what of my solutions I can present as OpenCL 2.0 conformant and this pertains to GCN2 and above. OpenCL is open, with conformance tests, and driver may support the much of it without achieving conformant status and/or they may also be supporting extensions that are part of OpenCL 2.0. I also don't personally know if the app is reading something (i.e. and extension) and assuming OpenCL 2.0 rather than the driver directly reporting something to the app.
 
Back
Top