Monday, June 10th 2019
AMD E3 Next Horizon Event: Live Blog
It's been a very busy May-June for AMD as the company pushes out its major client-segment product lines spread across Computex 2019, and E3 2019. At Computex, the company focused on its 3rd generation Ryzen "Zen 2" desktop processors, and led its partners to show us a galaxy of new motherboards based on the AMD X570 chipset. It turns out that the company was saving a handful processor SKUs focused on gamers for E3.
The second important product launch of course is Radeon RX 5700 series, based on AMD's new "Navi 10" silicon on which its new RDNA graphics architecture debuts. With its AIB (add-in board) partners expected to be allowed to make custom-design cards, and based on what little nuggets of information AMD put out, "Navi" promises to stir up a key performance-segment price-band that's currently held by NVIDIA's GeForce RTX 2070 and RTX 2060. The AMD keynote will see the company CEO Dr. Lisa Su and her top execs take centerstage to make some big announcements. With E3 being a purely entertainment / client-segment forum, the AMD keynote promises not bore with tiresome topics such as AI, self-driving cars, etc.
2:30 PM PDT: Ahead of its keynote, AMD posted a teaser video of its new RDNA graphics architecture on YouTube.
3:00 PM PDT: The event begins with CEO Dr. Lisa Su taking centerstage. "Exciting time for gamers everywhere, with coherence between hardware and software."
3:06 PM PDT: "AMD owes its recent success to big technology bets that are paying off now."
3:07 PM PDT: Lisa Su confirms next-generation PlayStation is powered by AMD IP, as is Google Stadia. Also talks about the Xbox "Project Scarlett"
3:08 PM PDT: PC. Now we're talking.
3:11 PM PDT: Key details on Zen 2 CPU architecture at the heart of Ryzen 3000. Also AMD's performance claims. We've seen these slides at the Computex 2019 keynote.
3:16 PM PDT: AMD answering Intel's "real-world gaming" challenge:
3:18 PM PDT: Ryzen 5 3600X specs confirmed, including gaming performance. Beats Core i5-9600K.
All CPU models available 7/7/2019.
3:20 PM PDT: Time to talk Navi. Maintains RDNA is a cleanslate architecture.
3:25 PM PDT: Meet the Radeon RX 5700 XT. Leadership performance in its class.
3:27 PM PDT: The leaks seem to be correct. "Ready for overclocking"
Performance beats the RTX 2070 in World War Z.
3:29 PM PDT: Radeon RX 5700 (non-XT) Specs:
Up to 10% faster than RTX 2060 (RX 5700 non-XT).
Radeon Media Engine supports 4K encode and decode, and updated codecs for hardware-acceleration. Radeon Display Engine has updated connectivity.
3:33 PM PDT: FidelityFX pre-baked free-to-use special effects that improve visual fidelity without performance cost; available through GPUOpen.
3:35 PM PDT: Radeon Image Sharpening is essentially FidelityFX for games that don't support it.
3:36 PM PDT: Radeon Anti-Lag is a groundbreaking new feature that works to reduce GPU-to-monitor lag and input lag. Treat for e-sports gamers.
Lower input-lag without increasing frame-rates!
3:40 PM PDT: Pricing! Availability 7/7/2019, both SKUs. RX 5700 XT is priced at $449, and RX 5700 at $379. "Gears 5" bundled!
3:47 PM PDT: Gears 5 as rendered on an RX 5700 XT.
Although Gears 5 releases September, RX 5700-series buyers who get the game bundled also have early access to pre-release testing versions from July.
3:54 PM PDT: Borderlands 3 is Ryzen+Radeon optimized, including FidelityFX. Gearbox also uttered "AMD Studios," an internal dev-relations moniker, detailed as a steady supply of AMD hardware to game developers at Gearbox. AMD is flexing its dev-relations muscle once again.
4:00 PM PDT: Unity talks about HDRP. Also confirms Ryzen+Radeon optimization, and support for FidelityFX.
4:07 PM PDT: Ubisoft confirms Ryzen+Radeon optimization for Tom Clancy's "Ghost Recon: Breakpoint," including FidelityFX, async-compute, FreeSync 2 HDR.
4:11 PM PDT: AMD launches Radeon RX 5700X 50th Anniversary Edition with 1.98 GHz Boost frequency, priced at $499, exclusively on AMD.com
4:12 AM PDT: AMD also announces Ryzen 9 3950X 16-core processor, priced $749, launches September.
The second important product launch of course is Radeon RX 5700 series, based on AMD's new "Navi 10" silicon on which its new RDNA graphics architecture debuts. With its AIB (add-in board) partners expected to be allowed to make custom-design cards, and based on what little nuggets of information AMD put out, "Navi" promises to stir up a key performance-segment price-band that's currently held by NVIDIA's GeForce RTX 2070 and RTX 2060. The AMD keynote will see the company CEO Dr. Lisa Su and her top execs take centerstage to make some big announcements. With E3 being a purely entertainment / client-segment forum, the AMD keynote promises not bore with tiresome topics such as AI, self-driving cars, etc.
3:00 PM PDT: The event begins with CEO Dr. Lisa Su taking centerstage. "Exciting time for gamers everywhere, with coherence between hardware and software."
3:08 PM PDT: PC. Now we're talking.
3:20 PM PDT: Time to talk Navi. Maintains RDNA is a cleanslate architecture.
3:33 PM PDT: FidelityFX pre-baked free-to-use special effects that improve visual fidelity without performance cost; available through GPUOpen.
3:54 PM PDT: Borderlands 3 is Ryzen+Radeon optimized, including FidelityFX. Gearbox also uttered "AMD Studios," an internal dev-relations moniker, detailed as a steady supply of AMD hardware to game developers at Gearbox. AMD is flexing its dev-relations muscle once again.
133 Comments on AMD E3 Next Horizon Event: Live Blog
For starters, they "caught up" with half a chip size (5700XT is 250mm^2, 2070 is 450mm^2).
On top of it, Huang complained that 7nm "is expensive", so hardly anything coming within next 6 month.
What is likely to come within 6 month is ngreedia's "super" bump and AMD's bigger Navi that could step well into 2080 area.
What 7nm gives over 12nm is 70-80% better transistor density, a 450mm^2 chip at 12nm would become 250mm^2 at 7nm. In December 2017 AMD stated 7nm was twice as expensive as 14/12nm, negating the effect of smaller die size.
The other big win from a smaller manufacturing process is better power efficiency and lower voltage accompanied by lower power consumption.
There is also a lot of room to grow performance for Zen3/Navi 2.0.
I think people will be surprised by how much more of a gain they can produce and I think it is a bit "cheeky" (I'm purposely not using the other word to avoid "reactions") to be making absolute statements that there is nothing more here (and in the future) other than node gains and that they will be behind in 6 months and other salty comments.
On top of that, there is actually a good pipeline of process improvements coming too.
But 450mm^2 14nm chip must cost more than twice more than 250mm^2 14nm chip.
AMD should have no problem beating 2070/2060 on price.
The exact division of R&D costs for all involved companies are unknown.
Nvidia outspends AMD by about 50% as a whole, it could be estimated 4-5 times for GPU division but whether that is true is debatable.
Intel outspends AMD about 10-fold for company as a whole but keep in mind that Intel does a lot more than AMD and some of these things are really expensive. Intel spends billions on foundries - TSMC is their main competitor in that field and TSMC should be a larger manufacturer than Intel is (and if I remember correctly Intel puts new fabs cost in R&D while TSMC does not). They work together with Micron on NAND/Optane/SSDs and Micron spends couple billion a year on R&D. There are more fields Intel competes in with further R&D costs and while they definitely have more to spend on CPU R&D the difference there is definitely not 10-fold but much less.
AMD spent about $1.5B in R&D in 2018.
Intel spent $13.5B, Nvidia spent $2.3B, TSMC spent $2.6B, Micron spent $2.3B.The problem is that we do not know how things stand right now.
I would expect 7nm cost to have been gone down but how much exactly we do not know. Pretty sure it is not even close to 12nm yet.
12nm is old, tried and true manufacturing process while 7nm is a new one. The yield dropoff with size increase is definitely much lower at 12nm and can be negligible at 450mm^2.
Related to this, one has to wonder why does Nvidia stay at 12nm? Their GPUs are power limited, the same as AMD's. Clocks for Nvidia GPUs are stuck at 2-2.1GHz on 12nm because we know after that point the power efficiency goes straight to hell. AMD is showing us that 2.0-2.1 is achievable on 7nm so that cannot be the problem. Yields and cost remain the main suspects here.