AMD Announces Radeon Pro 5600M Navi GPU with HBM2 - Inside Apple's MacBook Pro 16"
by Andrei Frumusanu on June 15, 2020 11:30 AM EST- Posted in
- Mobile
- AMD
- Laptops
- Radeon Pro
- Radeon Pro 5600M
Today AMD has officially announced one of the long rumoured missing Navi parts in the form of the new Radeon Pro 5600M mobile GPU, seeing the Navi 12 design finally take shape as a product.
The new high-end mobile GPU is a successor to the Radeon Pro Vega 20 and Vega 16 designs released back in 2018, products that ended up being used in Apple’s MacBook laptops. The new Radeon Pro 5600M also sees its debut in the new 16” MacBook Pro that’s also been debuted today. Apple has traditionally had exclusive rights to these mobile Radeon Pro SKUs so it’s likely this exclusivity also applies to the new Radeon Pro 5600M.
| AMD Radeon Series Mobile Specification Comparison | |||||||
| AMD Radeon Pro 5600M | AMD Radeon RX 5300M | AMD Radeon RX 5500M | AMD Radeon Vega Pro 20 | AMD Radeon RX 560X | |||
| CUs | 40 | 22 | 22 | 20 | 14/16 | ||
| Texture Units | 160 | 88 | 88 | 80 | 64 | ||
| ROPs | ? | 32 | 32 | 32 | 16 | ||
| Game Clock | N/A | 1181MHz | 1448MHz | N/A | N/A | ||
| Boost Clock | 1035MHz | 1445MHz | 1645MHz | 1300MHz | 1275MHz | ||
| Throughput (FP32) | 5.3 TFLOPs | 4.1 TFLOPs | 4.6 TFLOPs | 3.3 TFLOPs | 2.6 TFLOPs | ||
| Memory Clock | 1.54 Gbps HBM2 | 14 Gbps GDDR6 | 14 Gbps GDDR6 | 1.5 Gbps HBM2 | 7 Gbps GDDR5 | ||
| Memory Bus Width | 2048-bit | 96-bit | 128-bit | 1024-bit | 128-bit | ||
| Max VRAM | 8GB | 3GB | 4GB | 4GB | 4GB | ||
| Typical Board Power | 50W | ? | 85W | ? | ? | ||
| Architecture | RDNA (1) | RDNA (1) | RDNA (1) | Vega (GCN 5) |
GCN 4 | ||
| GPU | Navi 12 | Navi 14 | Navi 14 | Vega 12 | Polaris 11 | ||
| Launch Date | Q2 2020 | Q4 2019 | Q4 2019 | 10/2018 | 04/2018 | ||
The new mobile GPU is characterised by its large compute unit count as well as its usage of HBM2 memory. With a CU count of 40, resulting in 2560 stream processors, the Radeon Pro 5600M actually matches AMD’s current best desktop graphics designs such as the Navi 10-based Radeon 5700XT. A key difference here lies in the clocks, as this mobile variant only clocks up to a maximum of 1035MHz, resulting in a theoretical maximum throughput of 5.3TFLOPs, quite a bit less than its desktop counterpart which lands in at 9.75TFLOPs.
In terms of bandwidth however, the mobile chip more than keeps up with its desktop counterpart. AMD is using a 2048-bit HBM2 memory interface to up to 8GB of memory running at 1.54Gbps, resulting in a bandwidth of 394GB/s, only a bit less than the 448GB/s of the Radeon 5700XT.
The Radeon Pro 5600M is advertised with a total graphics power (TGP) of 50W, identical to the TGP of the Radeon Pro 5500M and the Radeon Pro 5300M. Both of those, in turn, are based on the Navi 14 die, which contains far fewer compute units. This makes the Radeon Pro 5600M an incredibly performant and efficient design – albeit one that's undoubtedly expensive to build.
The new Radeon Pro 5600M is now available inside of Apple’s MacBook Pro 16” as an BTO upgrade option, and comes at a $700 mark-up versus the default Radeon Pro 5500M GPU.
Facebook
Twitter
RSS
67 Comments
View All Comments
MarcusMo - Monday, June 15, 2020 - link
Likely content creators. Or anyone using gpu compute for that matter.PEJUman - Monday, June 15, 2020 - link
.... and just like that.. we're back to AI and machine learning. I tend to agree with Peter.... hard to see who is the target market for this. Thermally constrained Chassis and 50 W TDP with this kind of processing power = burst load instead of steady state continuous use of GPU based compute.So this rules out video encoding/content creators. AI load are occasionally transients; but at this price, can't argue the Amazon instance. So that leave the 3D modelers, which I think still mostly on windows ecosystem.
So Operandi, what would you use this workstation for?
Santoval - Tuesday, June 16, 2020 - link
"So that leave the 3D modelers, which I think still mostly on windows ecosystem."That depends on the 3D modeler / CGI artist, their workload, and their personal preferences. Many, in particular those who use Maya, Modo, Cinema 4D, Nuke, Arnold, Mudbox, Houdini and of course Blender, prefer the freedom, stability and flexibility of Linux.
Some software like 3Ds Max (one of the rare products of Autodesk, and arguably the most important, with no Linux port) and Lightwave 3D do not have a Linux port but they can export files in commonly used formats that can then be imported by some other software running on Linux to continue work there. It is a hassle though, since it requires a separate computer (either physical or virtual), so most artists who work on Linux tend to avoid it.
As a rule, the higher the budget of a film (or TV show) the more likely it is for most or all of its CGI to have been done on Linux. The most premier VFX houses that work on blockbuster films tend to avoid Windows.
PEJUman - Tuesday, June 16, 2020 - link
How about a TB3 eGPU for this application? Which one would be a better $700 investment?I am asking this from the typical usage of a desk + docking station and the laptop being portable only for meeting/customer pitches. i.e. how often do you need to this done on your workstation laptop/on-site @ client location?
vFunct - Tuesday, June 16, 2020 - link
People hereat like photographers & videographers are a tiny market.There are a lot more content creators than there are AI researchers.
vFunct - Tuesday, June 16, 2020 - link
^actPEJUman - Tuesday, June 16, 2020 - link
I am under the following assumptions for adobe consumer:These GPUs does not scale with CUs under typical adobe load. It scales more with TDP limits... which makes spending $700 for more CUs but the same TDP makes the 5600M only better in some corner cases (i.e. brief spike to full load, when the chassis thermal mass + heat rejection capability allows the 5600M to stretch its legs for 1-2 minutes).
Am I incorrect in this assumption?
AlexDaum - Wednesday, June 17, 2020 - link
@PEJUman, most GPU compute applications scale very well with CU count, so it probably will have better performance. And more CUs at lower clock Speed consume less power, so sustained performance in thermally constrained package.Samus - Tuesday, June 16, 2020 - link
True, a lot of Adobe plugins use GPU resources.skavi - Monday, June 15, 2020 - link
it is no worse than any computer of comparable size, with similar battery capacity. better than most of that class in fact.