Qualcomm Announces Snapdragon 8+ Gen 1: Switch to TSMC for More Speed, Less Power

As the darkness of night arrives in China tonight, Qualcomm is hosting a mobile-focused product launch event they’re calling “Snapdragon Night.” Headlining the event was the announcement of the company’s new flagship SoC, the Snapdragon 8+ Gen 1. A mid-generation update to their flagship smartphone SoC, the Snapdragon 8 Gen 1, the 8+ Gen 1 follows Qualcomm’s annual tradition of releasing a product refresh to boost performance and give partners something new to work with for the second half. And for this year in particular, we’re looking at a very noticeable change in chips from Qualcomm.

Unlike previous generations where Qualcomm simply launched a faster speed bin of their existing silicon, for 2022 we have something more substantial to discuss. Qualcomm has switched foundries entirely – from Samsung to TSMC – and, as a result, is rolling out a new matrix. Thanks to this, the Snapdragon 8+ Gen 1 Qualcomm reaps something of a unique manufacturing gain, allowing them to both increase CPU and GPU performance while simultaneously reducing power consumption.

Flagship Qualcomm Snapdragon 8 Gen 1 SoC
SoCs Snapdragon 8+ gen 1 Snapdragon 8 Gen 1
CPU 1x Cortex-X2
@ 3.2GHz

3x Cortex-A710
@ 2.8GHz

4x Cortex-A510
@ 2.0GHz

6 MB sL3

1x Cortex-X2
at 3.0GHz

3x Cortex-A710
at 2.5GHz

4x Cortex-A510
at 1.8GHz

6 MB sL3

GPUs Adrene
(10% higher clock speed)
Adrene
DSP / NPU hexagon hexagon
Memory
Controller
4 channels 16 bit

@ 3200 MHz LPDDR5 / 51.2 GB/s

4 MB system level cache

ISP/Camera Triple Spectra 18-bit ISP

1x 200MP or 108MP with ZSL
or
64+36MP with ZSL
or
3x 36MP with ZSL

8K HDR video and 64MP burst capture

Encode/
Decode
8K30/4K120 H.265 10bit

Dolby Vision, HDR10+, HDR10, HLG

720p960 infinite recording

Built-in modem Integrated X65

(5G NR Sub-6 + mmWave)

DL = 10000 Mbit/s

UL = 3000 Mbit/s

MFC. Process TSMC 4nm Samsung 4nm

Quickly diving into the specs, the new Snapdragon 8+ Gen 1 is essentially the original Snapdragon 8 Gen 1 transferred from Samsung’s 4nm line to one of TSMC’s 4nm lines. Under more normal circumstances, this kind of change would probably be trivial – or at most, a fun exercise in finding edge cases – but for Qualcomm’s flagship SoC, the question is more important.

While official sources and statements about the quality of Samsung’s 4nm process are scarce, unofficially it has become clear that Samsung’s 4nm process has failed to meet expectations. This had a cascading impact on the chips made on the process node, leading the original Snapdragon 8 Gen 1 to develop an affinity for power consumption, and Samsung’s Exynos 2200 did not. doesn’t come out any better. Conversely, by all accounts, TSMC’s N4 process looks great, with the optically shrunken node building on TSMC’s already successful and highly capable 5nm technologies.

Because of this performance gap between Samsung’s and TSMC’s 4nm nodes, Qualcomm is taking the unusual step of moving (essentially) its high-end SoC to TSMC’s factory. Which, while not strictly necessary – Qualcomm has plenty of momentum and the 8 Gen 1 is selling well – is certainly a prudent move for the company. Qualcomm faces particularly stiff competition this generation from MediaTek, whose flagship-level SoC Dimensity 9000 was the flagship of TSMC’s 4nm node. And that leaves MediaTek with a distinct advantage over the original 8 Gen 1, which Qualcomm would be very happy to negate.

Ultimately, the change in fabs gives Qualcomm a chance to improve on the original 8 Gen 1 from both ends of the spectrum, resulting in the Snapdragon 8+ Gen 1. Performance-wise, TSMC’s Node gives them an easy opportunity to boost CPU and GPU clock speeds for more performance. The main Cortex-X2 core is now clocked 7% higher at 3.2GHz, and meanwhile the A710 and A510 clusters have seen their clock speeds increase even more significantly, by around 12% each. Now even the slowest A510 cores can run at 2 GHz. GPU clock speeds have also been similarly increased, and while Qualcomm doesn’t disclose specific clock speeds, they have confirmed that the 8+ Gen 1’s Adreno GPU block is clocked 10% higher than the original 8 Gen 1.

But, if anything, the bulk of Qualcomm’s gains from switching manufacturing nodes is invested in reducing power consumption. Something of a sore point with 8 Gen 1, TSMC’s better 4nm process means Qualcomm is seeing much lower power consumption on its iso-frequency SoC.

Officially, Qualcomm claims a 30% improvement in GPU and CPU power efficiency. Although, as mentioned earlier, this is an iso frequency and does not take into account the higher peak clock speeds of the 8+ Gen 1. Therefore, the actual power savings will not be as significant at peak times. on a peak basis, but according to Qualcomm, energy savings are still significant. Overall, the company is touting a 15% reduction in SoC power consumption under “practical usage models” compared to the original 8 Gen 1, which in turn should improve the battery life of handsets that adopt the new SoC.

Beyond that, the official 8+ Gen 1 specs don’t reflect any significant SoC configuration changes from the original chip. So we’re still looking at the same built-in X65 5G modem, same Spectra ISP, and same video encode/decode blocks (sorry, gang, still no AV1 support!). So despite assembling a new die for their mid-gen refresh product this year, there are no new features to speak of with the 8+ Gen 1.

When it comes to SoC performance, Qualcomm officially claims a 10% improvement in GPU and CPU performance, due to the aforementioned clock speed increases. While we weren’t able to attend a benchmarking session hosted by Qualcomm last week, the performance mode numbers released by the company are roughly in line with these claims. Qualcomm’s Geekbench 5 results are several percent higher than those we compared in December at the 8 Gen 1 launch event, although it’s notable that they didn’t score significantly higher in PCMark. GPU performance numbers are also mixed, with some of Qualcomm’s official results coming very close to our original 8 Gen 1 results, but I’m hesitant to read too much given the difference between a peak and sustained tests in the GPU results. As always, the final say will have to come down to independent third-party testing, although on the face of it nothing that Qualcomm claims is unreasonable given the clock speed and thermal headroom improvements achieved by moving to TSMC.

Finally, as for consumers, they will be able to get their first glimpse of Snapdragon 8+ Gen 1 devices in the third quarter of this year. According to Qualcomm, many of the usual suspects have pledged to launch phones based on the new SoC, including Asus, Motorola, OnePlus, Honor and Xiaomi.

#Qualcomm #Announces #Snapdragon #Gen #Switch #TSMC #Speed #Power