Samsung Announces the Exynos 9825 SoC: First 7nm EUV Silicon Chip
by Andrei Frumusanu on August 6, 2019 9:30 PM EST- Posted in
- Mobile
- Samsung
- Smartphones
- Exynos
- SoCs
- Exynos 9825
This year’s Galaxy S10 has been in a bit of an odd situation: Although Samsung continued to dual-source its SoCs, using both its own Exynos 9820 SoC as well as Qualcomm’s Snapdragon 855, the phone found itself in the unusual situation of pitting 8nm silicon against 7nm silicon from TSMC. So although the new Exynos 9820 did fairly well in testing and improved a lot over the Exynos 9810, the chip seemingly still had disadvantages against the competition when it came to power efficiency, likely linked to its process technology disadvantages. On top of the power efficiency disadvantages, the chip also had a notable die area disadvantage versus the Snapdragon, coming in at 127mm² versus the smaller 73mm² competition.
Samsung’s 7nm EUV process node was noted as having started production back in October of last year, although we’re not sure exactly which chip this was referring to, and we had hopes that it would be the chip for the S10 but alas it was not to be.
This time around, Samsung is seemingly bridging the gap with the introduction of the new Exynos 9825 – a 7nm LPP refresh of the Exynos 9820.
Samsung Exynos SoCs Specifications | |||
SoC |
Exynos 9820 |
Exynos 9825 |
|
CPU | 2x M4 @ 2.73 GHz 2x 512KB pL2 2x Cortex A75 @ 2.31 GHz 2x 256KB pL2 4x Cortex A55 @ 1.95 GHz No pL2's Shared complex sL3 @ 4MB |
2x M4 @ 2.73 GHz 2x Cortex A75 @ 2.4 GHz 4x Cortex A55 @ 1.95 GHz |
|
GPU | Mali G76MP12 @ 702 MHz | Mali G76MP12 @ ? MHz | |
Memory Controller |
4x 16-bit CH LPDDR4X @ 2093MHz |
4x 16-bit CH LPDDR4X @ 2093MHz |
|
ISP | Rear: 22MP Front: 22MP Dual: 16MP+16MP |
Rear: 22MP Front: 22MP Dual: 16MP+16MP |
|
Media | 8K30 & 4K150 encode & decode H.265/HEVC, H.264, VP9 |
8K30 & 4K150 encode & decode H.265/HEVC, H.264, VP9 |
|
Integrated Modem | Shannon 5000 Integrated LTE (Category 20/13) DL = 2000 Mbps 8x20MHz CA, 256-QAM UL = 316 Mbps 3x20MHz CA, 256-QAM |
Shannon 5000 Integrated LTE (Category 20/13) DL = 2000 Mbps 8x20MHz CA, 256-QAM UL = 316 Mbps 3x20MHz CA, 256-QAM |
|
Mfc. Process | Samsung 8nm LPP |
Samsung 7nm LPP (EUV) |
The new chip very much looks like a die-shrink/mid-cycle refresh with largely the same IP generation as the 9820, still featuring Samsung’s M4 Cheetah cores as well as a Mali-G76 GPU. Samsung also doesn’t seem to have changed the clock frequencies of the chip very much: The M4 cores are still running at a peak frequency of 2.73GHz and the A55 cores also run at 1.95GHz. We do see a bump in the frequencies of the middle cores that goes up from 2.31GHz to 2.4GHz.
On the GPU side, Samsung has also stuck with the same GPU configuration as with 9820, using a MP12 configuration of the G76. According to the company the 9825's GPU is clocked higher - so it will outperform its predecessor - however the company has yet to disclose specific clockspeeds.
As for the integrated modem, Samsung has retained their Shannon 5000, a Category 20/13 modem. This modem has a peak download rate of 2 Gbps (with 8x carrier aggregration), while uploads top out at 316 Mbps. We had been wondering if Samsung would be able to squeeze in a 5G modem for this SoC, but it looks like it's just a bit too early for that. Instead, 5G can be accomplished by pairing the SoC with Samsung's 5G Exynos Modem 5100.
The new chip is likely to be featured in the new Galaxy Note10 – Samsung will continue to use Snapdragon chips for some markets, and this could be an explanation for the new chip not having that big improvements on the part of the CPU complex as it’s aiming for performance parity with the Snapdragon. We also have to note that Samsung would have to invest the process improvements into improving power efficiency rather than raising performance.
The chip reminds us of the Exynos 5430 from a few years ago which was also a process-shrink to the chip that ended up in the Galaxy S5, representing Samsung’s first 20nm silicon. That chip never ended up in the popular flagship products, but seemingly did serve a purpose as a pipe-cleaner and learning platform for the new process node. The new Exynos 9825 could end up in a similar situation, although being used in the Note10, it won’t nearly have an as long lifespan as we don’t expect it to power the Galaxy S11 next year.
Source: Samsung
49 Comments
View All Comments
Dragonstongue - Wednesday, August 7, 2019 - link
competition from the same company to the same company (Samsung/TSMC) is not so much competition now is it?unfortunately at this "high level" there are so very very few players that actually hold any cards worth playing.
also when was last time you heard of a major player doing an across the board "lower price" ... the last company I personally heard of doing so making almost a mantra was AMD with Ryzen "more cores for less $$"
But as far as Samsung to be a good second source to TSMC, ummmm, yeh, maybe, seeing as Samsung/TSMC/AMD/IBM and a few others are "buddy buddy" chasing the last bit of life silicon has to it, maybe the way they are rushing the so called die "shrinks" TSMC, Samsung etc will be able to go back to 16nm or so and "do it again" to get a true 100% die shrink, not the mix of some stuff gets shrunk and the rest does not.
anyways....Samsung orders from TSMC, last I heard TSMC does not get anyone to do their FAB work, seeing as THEY do the fab work for others
LOL
levizx - Wednesday, August 7, 2019 - link
What a stupid comment! TSMC is a PUREPLAY fab. Samsung Group is NOT. Samsung Foundry does not order from TSMC, as THEY do the fab for others.FullmetalTitan - Thursday, August 8, 2019 - link
This might help you learn something:https://www.quora.com/What-does-pure-play-or-IDM-m...
Yojimbo - Wednesday, August 7, 2019 - link
You think Intel will contract Samsung to produce chips for them?! Any source for that or is it just your own speculation?name99 - Wednesday, August 7, 2019 - link
Why not? Once again, even though the internet seems to think semiconductors is a sports game based on cheering and kayfabe, in the actual world businesses co-operate all the time.Intel, for example, bought Infineon. And manufactured modems on TSMC for a few years.
Intel bought Movidius. And manufactured vision chips on TSMC (I think all Movidius products being sold by Intel are STILL on TSMC, already through one upgrade cycle, and have no idea if they plan to stick with that or change it).
Gondalf - Wednesday, August 7, 2019 - link
Not only, Stratix apart, nearly all the Intel Altera production is on TSMC, moreover Intel Nervana chip is on TSMC process too.Gondalf - Wednesday, August 7, 2019 - link
Only Logic?? sometimes speculation is useless. A GPU need of a lot of silicon footprint, that silicon not necessary is of very high value because clocks are low and the main problem is to keep leakege down lowering the voltage.Why to utilize Intel Fabs for this silicon area?? We don't need of high levels of drive current, far better manufacture GPUs in a Foundry. Internal Fabs are better suited to print server dies and high margins consumer SKUs.
Anymoore - Wednesday, August 7, 2019 - link
Something important missing: https://www.mitsuichem.com/en/release/2019/2019_05...NICOXIS - Wednesday, August 7, 2019 - link
This die shrink should put it on par with 855 nowAnymoore - Wednesday, August 7, 2019 - link
Applying 40% area reduction (127 * 0.6), would almost catch up.