• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

AMD Radeon RX 590 GME is a Dressed Up RX 580: No more 12nm, Lower Performance

AMD promised 2020 to be a better year. So far, 2020 turns out to be a disastrous year for everyone.
3 months on in 2020, and AMD hasn't released anything significant yet.

AMD CEO Lisa Su interview — 2020 will be a bigger product year for us

2020 Will Be High Performance Computing Year for AMD, Says Lisa Su


AMD Zen 3 processors will arrive by March 2021 at the latest
And 5nm AMD Zen 4 will follow by 2022

N5 Zen 4 in 2022, while no word on N5 GPUs ?
 
Eh, 280 was a rebranded first gen GCN 79xx, 285 (which was later rebranded as the 380) was a new 3rd gen GCN chip more like Fiji (FuryX) though. Meanwhile you had second gen GCN with Hawaii as the 290 and later rebranded as the 390. AMD's lineup is/was messy.

Also funnily enough, Tonga (285, 380) has the highest performance and still to date (even beating out Navi) best quality H.264 hardware encoding from AMD.

The R9 285 which became the R9 380/380X also ditched the 384bit memory bus on the 280/79xx.

The R9 285/380 is actually a 28nm RX 480.

The Fury was a 28nm Vega 64 but it had the crappy FPU that underpins Polaris and is bugged in Navi. Vega took the FPU from the 290/390 and that's where the split really started, with a gaming core and a pro core.

The Fury was supposed to replace the 290/390 for gamers. Sadly HBM turned out to be a hassle and the massive core had problems, which driver optimization helped sort until AMD kind of ditched HBM optimizations, until they had to pull Vega back into the mainstream.

The Fury benefits from undervolting, as hot spots do affect the core and impact the speed.

The GeForce Style GCN family tree:

Fury X: 64 CU
285, 380, 480, 580, 590: 32CU

Of course the design was process shrink and tuned up but the GFX engine major version remained the same.

285/380: 28nm

Graphics/Compute: GFX8 (gfx802)
Display Core Engine: 10.0

Fury X: 28nm

Graphics/Compute: GFX8 (gfx803)
Display Core Engine: 10.0

590: 12nm

Graphics/Compute: GFX8 (gfx803)
Display Core Engine: 11.2
 
Last edited:
Ellesmere/Polaris 10/20/30 has 36 CUs.

Tonga ships with only 28 CUs enabled, if it has more, they were used only in hidden cases (for instance for Apple orders, or something).


AMD gets the blame from videocardz :)

This is without a doubt a continuation of misleading product naming by AMD

AMD’s Polaris 20 is a 14nm processor, while Polaris 30 is 12nm. The latter offers more headroom for higher clocks and there is no logical reason to call the new model an RX 590. Yet, here we are.

The conclusion of the review is straightforward: Radeon RX 5500 XT is a much better product, although it costs 1399 Yuans, while the RX 590 GME retails at 1199. Not to mention, the Navi 14-based model offers more OC headroom and consumes less power.

Here we go, guys. Enough said :)
 
Last edited:
Yeah the 285 = 380 which was a cut down 380X was more in line with the 470/570 that's a cut down 480/580.

285: 1792 / 112 / 32 - 28CU
290X: 2816 / 176 / 64 - 44CU
380: 1792 / 112 / 32 - 28CU
380X: 2048 / 128 / 32 - 32CU
Fury X: 4096 / 256 / 64 - 64 CU
470: 2048 / 128 / 32 - 32CU
480: 2304 / 144 / 32 - 36CU
Vega 10: 4096 / 256 / 64 - 64 CU
580: 2304 / 144 / 32 - 36 CU

Scorpio: 2560 / 160 / 32 - 40CU

Navi 10: 2560 / 160 / 64 - 40CU

Polaris was bump up in pipe design that nicely cut down to match the 380X. Bolt on upgrades get cut down.

Vega took the basic layout of the Fury, upgraded it and gave it much better FPU capabilities like the 290/390.

Scorpio is the One X GPU and it's based on the 290X and crushes Polaris under it's 384bit memory bus boot. Also suspiciously looks like Navi's layout... Well hey now you know why Navi is bandwidth limited with 256bit memory... LoL

Transistor numbers:

290X: 6,200 million
380X: 5,000 million
Fury X: 8,900 million
480: 5,700 million
Scorpio: 7,000 million
Vega 10: 12,500 million
Vega 20: 13,230 million
Navi 10: 10,300 million

In FP64 Perf

380X = 470 at around 300Gflops
Fury X = 590 at around 500Gflops
5700XT 50th AE sits around 600Gflops
390X = V64 LC at around 800Gflops
Vega 20 sits around 7,400 Gflops(commercial versions, consumer is 50%)
 
Last edited:
Sounds like a card for the extreme budget builders, and Venezuela...
 
another misleading GPU model because releasing a RX580 with less cores than the original wasn't enough...
 
RX 5100 or RX 5200.



What does "they are making chips" even mean?
The Chinese will buy RX 5500 or RX 5500 XT instead of this.

This is old, weak, power hungry, expensive, lacks any new modern features and qualities.
It is basically a chip based on R&D in 2012-2013.

Meh, move on...

TSMC 7nm is also running at 100% capacity. AMD is selling every Navi and Zen2 chip they can make, so there's no choice but to keep producing 14nm chips so they can make additional money. Same thing with Intel still manufacturing 22nm chips because their 14nm+++ node is completely occupied.
 
Eh, 280 was a rebranded first gen GCN 79xx, 285 (which was later rebranded as the 380) was a new 3rd gen GCN chip more like Fiji (FuryX) though. Meanwhile you had second gen GCN with Hawaii as the 290 and later rebranded as the 390. AMD's lineup is/was messy.

Also funnily enough, Tonga (285, 380) has the highest performance and still to date (even beating out Navi) best quality H.264 hardware encoding from AMD.

Sure that the ancient 285 or 380 was the most powerefficient one? i would say the 7970 or which series held up against its younger version for sooo long?
 
Honestly not a fan of rebrands. I'm one of those who likes to play with the newest toys, and the possibility of accidentally buying something last gen sucks
 
Last edited:
Source? I love how people make facts out of thin air.
TSMC 7nm production capacity cant handle all its orders - GizChina
TSMC 7nm chip supply remains tight - Digitimes
TSMC Announces Major Expansion to Feed 7nm, 5nm Demand - ExtremeTech

CPU Shortages Might Be Set to Take a Toll on PC Industry Once Again - TheStreet

There you go. TSMC is actually expanding just to fill in demand so like I said 7nm is supply constrained. AMD has to continue producing 12 and 14nm chips as alternate sources of revenue, just as Intel actually cancelled EOL for some 22nm chips.
 
TSMC 7nm production capacity cant handle all its orders - GizChina
TSMC 7nm chip supply remains tight - Digitimes
TSMC Announces Major Expansion to Feed 7nm, 5nm Demand - ExtremeTech

CPU Shortages Might Be Set to Take a Toll on PC Industry Once Again - TheStreet

There you go. TSMC is actually expanding just to fill in demand so like I said 7nm is supply constrained. AMD has to continue producing 12 and 14nm chips as alternate sources of revenue, just as Intel actually cancelled EOL for some 22nm chips.

These are old news and fake for the current situation - declining PC sales.
And I asked you for a proof that AMD sells every single chip they make, which of course can't be true.

Report: GPU, Motherboard Shipments Hitting Record Lows in Wake of Coronavirus Outbreak
 
Back
Top