Wednesday, February 25, 2015

Early benchmarks for MT6795 show high performance, suggest use of eight Cortex-A53 cores

MediaTek originally announced the MT6795, a SoC targeting the premium-level and performance segments of the smartphone market, in July 2014, with expectations of devices being commercially available to end users before the end of 2014. However, the chip was delayed (problems with the memory controller were reported) and competitive benchmark results are only now beginning to surface for the chip.

According to the announcement, the SoC was to have an octa-core CPU configuration with clock speeds up to 2.2 GHz, a strong dual-channel memory interface with support for LPDDR3 up to 933 MHz, 2K (2560x1600) display support. Other reports and information have suggested that it uses a PowerVR G6200 GPU, similar to the one used in MediaTek's MT6595, which can be seen as 32-bit predecessor of the new chip.

Confusion about processor cores, octa-core Cortex-A53 seems likely

The actual CPU cores used inside the MT6795 continue to be source of confusion. Initially understood to be an octa-core Cortex-A53 CPU configuration clocked at a high frequency, later a purported leaked MediaTek product roadmap surfaced that described the MT6795 as a big.LITTLE design that includes Cortex-A57 cores. However, a recent new entry in the Geekbench database suggesst that the chip actually has eight Cortex-A53 cores as originally suspected, as the IPC (instructions per cycle) of the integer and floating point subtests would be hard to reconcile with Cortex-A57 cores being present.

Geekbench results show mixed performance but high overall score

The Geekbench results show strong CPU performance, with the overall score being superior to that of available results for Snapdragon 810, which has a significantly higher cost design but has been plagued by performance issues, although it scores lower than Exynos 5433/Exynos 7 Octa with Cortex-A57 cores as used in the Galaxy Note 4. Note that MT6795 uses a less advanced 28 nm process compared to the 20 nm process used for Snapdragon 810 and Exynos 5433.

Single-score integer performance is not spectacular and below that of the previous generation high-end chips such as Snapdragon 801. Although this is compatible with the use of medium-performance Cortex-A53 cores, integer single-core performance is actually lower than the mid-range MT6752, despite the higher clock rate, pointing to continuing hardware performance problems with the chip. The Dijkstra benchmark result is particular low. This benchmark has a lot of external memory access and likely branches a lot, taxing certain elements of the CPU and SoC that simpler CPU benchmarks do not. It may be affected by the doubled address size in AArch64 mode, either through the increased size of pointer storage or reduced efficiency of the branch prediction unit inside the processor core.

Single core floating point performance in the Mandelbrot benchmark is higher than the MT6752 and actually compatible with the Cortex-A53 core running at 2.1 GHz, close to the originally envisaged maximum clock speed for the MT6795. Multi-core performance in this subtest is impressive, with a score that is higher than most existing SoCs including Exynos 7 Octa, which employs faster Cortex-A57 cores.

Finally, the dual-channel memory interface seems to working reasonably well in the tested revision of the chip/development board, with memory scores consistent with an optimized dual-channel interface, and higher, for example, than those of Exynos 5433. However, they are generally lower than those of the 32-bit MT6595.

One caveat is that the MT6795 entry is running in AArch64 mode, while the other devices were running in AArch32 (32-bit ARMv8) or 32-bit ARMv7 mode.

Average single-core CPU performance, strong multi-core performance

In a direct comparison with the MT6752, which has a comparable CPU configuration but clocked lower and has only a 32-bit memory interface, the MT6795 is only slightly faster, although the MT6795 uses a full 64-bit AArch64 instruction set model, while the tested MT6752 configurations use AArch32 with partial use of ARMv8 features. There are a few anomalous results, including a low score for the MT6795 in the single-core AES benchmark, and as mentioned it also scores significantly lower in the Dijkstra benchmark. Floating point performance is consistently higher for the MT6795 (more than the increase in clock rate would explain), which may be caused by the higher-performance memory subsystem of the MT6795 and/or the increased number of floating point registers available in AArch64 mode.

The MT6795 is clearly slower than its 32-bit predecessor MT6595 (which uses high-performance Cortex-A17 and Cortex-A7 cores in a big.LITTLE configuration) in most metrics, with only the heavy weighting and large performance gain for the AES and SHA1 cryptography tests  (due to the new ARMv8 instruction set) shifting the advantage for the overall score towards the MT6795.

When making a comparison with a median entry for the high performance Exynos 5433 (Exynos 7 Octa) inside the Samsung Galaxy Note 4, the MT6795 fairly consistently shows clearly lower single-core performance but higher multi-core performance.

MT6795 likely to be most cost-effective performance segment processor on the market

The exclusive use of Cortex-A53 CPU cores, and not the much more expensive and die-space consuming Cortex-A57 (or, in a 32-bit comparison, Cortex-A15/A17 cores), has positive implications for the cost of the chip. Die space dedicated to the CPU cores will be relatively low, although L2 caches will take considerable space when configured with a size that matches the desired performance level and market segment. Overall, the chip is likely to be attractive in terms of performance/dollar for the performance segment.

In terms of SoC optimizations, the chip would probably work better with the employment of additional ARM IP such as a Mali T760 or Mali-T800 series GPU, which offers advantages in combination with ARM cores such as Cortex-A53 in tandem with techniques such as AFBC, smart composition and transaction elimination, and new interconnect buses within the chip. SoCs like the MT6752 probably benefit from these optimizations, while the MT6795 cannot do so fully because of the non-ARM GPU. It seems likely that the MT6795 will be superseeded in next generation products to be announced by MediaTek in the future by a similar SoC with an ARM Mali-T760 or T800 series GPU.

Update (2 March): Based on a closed-door presentation event at the MWC, MediaTek appears to have rebranded MT6795 as Helio X10 with future Helio P series products also being announced.

Sources: MediaTek (MT6795 announcement), Geekbench browser

Qualcomm has announced new SoCs, uses new Cortex-A72 core

Recently, Qualcomm announced a number of new SoCs for the cost-sensitive and performance segments of the smartphone market, namely Snapdragon 415 and Snapdragon 425 in the 400 series, and Snapdragon 620 and Snapdragon 618 in the 600 series.

New Snapdragon 415 an 425 offer mid-range performance features

Qualcomm's product line has been somewhat inconsistent recently, with products from a series for a certain segment actually being used for a different segment. For example, although the Snapdragon 410 SoC is the mid-range 400 series, it has actually been deployed in significant numbers of cost-sensitive entry-level 4G segment devices.

There used to be a gap in Qualcomm's product line, large in terms of performance level, between the lower mid-range Snapdragon 400 and the premium level Snapdragon 801. Not too long ago, Qualcomm addressed this gap with the mid-range Snapdragon 615, featuring a total of eight Cortex-A53 cores, four with maximum frequencies in the 1.5 - 1.7 GHz frequency range and four clocked lower (e.g. 1 GHz) for lower consumption. With the new Snapdragon 415, Qualcomm is bringing a SoC similar to Snapdragon 615 to the cost-sensitive mid-range segment, largely replacing the Snapdragon 410 for that tier (as I have discussed previously, Snapdragon 410's performance is flawed in several ways).

Snapdragon 415 could be a rebranding of 615 to replace 410, or maybe not

In fact, there is a possibility that Snapdragon 415 is actually the same chip and in fact a rebranding of the same product. Both Snapdragon 415 and Snapdragon 615 have a roughly similar CPU set-up (eight Cortex-A53 cores), an identical GPU (Adreno 405) and a Cat 4 LTE modem. Although Qualcomm in its press release mentions commercial availability in end-user devices for new chips will happen the second half of the year,  if it is the same chip it is likely that Snapdragon 415 will appear earlier (since it has essentially already in production for some time as Snapdragon 615), replacing Snapdragon 410. However, in its specifications page for Snapdragon 415, Qualcomm does not mention any distinction in CPU speed between cores, making it likely that it can run all cores at the maximum clock frequency, similar to MediaTek chips already on the market.

Meanwhile, Snapdragon 425 has a CPU configuration similar to Snapdragon 415 with a higher maximum clock speed, and also the same GPU, but has a more advanced modem with Cat 7 LTE, and better ISP functionality for camera processing. A comparison can be made with MediaTek's MT6752 which also has a 1.7 GHz octa-core Cortex-A53 CPU (Snapdragon 425 probably also drops the pseudo-big.LITTLE design of Snapdragon 615). Given the clock speeds, it is likely that Snapdragon 425 is manufactured on a higher performance process than TSMC's 28LP, most likely TSMC's 28HPM, like MediaTek's chips.

Symmetric octa-core CPU configuration has advantages for multi-threaded applications

Although "LITTLE" cores in a big.LITTLE configuration can be taken advantage of in multi-threaded algorithms, most applications and algorithms are designed for and work best with processor cores running at a comparable speed, distributing the workload evenly between cores, favouring symmetric CPU configurations in which every core can run at the same maximum frequency. This shows in the very high multi-core benchmark scores of chips using such a configuration, such as MT6752. It looks like Qualcomm is quickly moving towards such as symmetrical octa-core configuration (pioneered by MediaTek, which already has comparable chips on the market) for the cost-sensitive part of the market, up to the mid-range segment.

Snapdragon 415 and 425 not likely to be cheap in terms of manufacturing cost

Although the eight Cortex-A53 cores are relatively small so their consumption of die space is relatively limited, as I discussed earlier the Adreno 405 GPU, with its medium-level performance, appears to have characteristics of a GPU targeted at higher-end segments (in terms of ALU/shader performance, for example) and is likely have a relatively large die size in relation to the cost-sensitive segments it is addressing. Because of that, Snapdragon 415 seems to be somewhat of a stop-gap measure to replace the successful, but flawed in terms of performance, Snapdragon 410 SoC, as the gross margin on this chip could be relatively small.

The proliferation of chips such as Snapdragon 415 likely to continue Qualcomm's heavy reliance on TSMC's 28LP process, which is lower-performance process technology than 28HPM. Why Qualcomm would place such emphasis on this process for smartphone SoCs is unclear, since the advantages of the 28/20HPM process are very desirable for smartphone SoCs for everything but the entry-level segment, and competitor MediaTek has adopted this process for most of its range. Qualcomm has been using 28HPM and 20HPM for its Snapdragon 801, 805 and 810, although the it is likely Snapdragon 425, 618 and 620 will also be using it.

Little heard from quad-core Cortex-A53 Snapdragon 610

It would have made sense if the Snapdragon 610 (announced as quad-core Cortex-A53 CPU and the same Adreno 405 GPU as the products discussed above) would have trickled down to the 400 series. The fact that this chip has barely appeared on the market and that it is not mentioned in the press release suggests it won't come to market at all, perhaps due to technical problems with the chip or as the result of a strategic decision. An updated quad-core Cortex-A53-based solution would certainly make sense in Qualcomm's product line.

Snapdragon 618 and 620 have premium-level characteristics

Qualcomm also announced two new performance segment processors, Snapdragon 618 and 620. These are the first announced mobile chips to feature the new Cortex-A72 processor core from ARM, which is an improved version of the high-performance Cortex-A57 processor core. Snapdragon 620 has four Cortex-A72 CPU cores clocked up to 1.8 GHz  and four Cortex-A53 cores up to 1.2 GHz in a big.LITTLE configuration, while Snapdragon 618 reduces the number of Cortex-A72 core to two to provide a better balance in terms of cost.

Although on the surface the model numbers of these new SoCs may seem close to Snapdragon 615, their specifications suggest that they are targeting a significantly higher performance segment. The memory interface is a dual-channel interface supporting LPDDR3 up to 933 MHz, clearly a defining feature for a high-end product, and making the support for QHD (2560x1600) displays a sensible feature. They also feature a new, "next-generation" GPU.

In short, despite their model number, Snapdragon 618 and 620 have little to do with Snapdragon 615 and should be thought of as processors in the same segment as processors from the Snapdragon 800 series such as as the Snapdragon 801 and Snapdragon 808. If and when Snapdragon 808 (with two Cortex-A57 cores and four Cortex-A53 cores) will appear on the market is unclear (some test results have appeared in the Geekbench database), the new announcement might suggest that it will quickly be superseeded by Snapdragon 618.

Sources: Qualcomm (SoCs announcement)

Updated February 26, 2015 (Edited and expanded to reflected likelyhood that Snapdragon 415 and 425 use a symmetrical CPU configuration, not pseudo-big.LITTLE like in Snapdragon 615).

Tuesday, February 17, 2015

Cortex-A53 not as power efficient as Cortex-A7

Recent detailed technical review articles published by AnandTech based on a comparison of Samsung Exynos SoCs have elucidated some of the details about the performance of the Cortex-A53 core, including processing performance, power consumption and die size. Overall, it appears that while Cortex-A53 is significantly faster than Cortex-A7 at the same clock speed, die size and power consumption on an equivalent manufacturing process has increased by a greater amount, leading to lower performance/Watt.

Direct comparison of Cortex-A7 and Cortex-A53 on the same process

In a recently published technical review article about the ARM Cortex-A53, Cortex-A57 CPU cores and Mali-T760 GPU core, based Samsung's Exynos-based Galaxy Note 4 model, AnandTech has provided details about the performance, power consumption and die size of the 64-bit Cortex-A53 core relative the its 32-bit predecessor, Cortex-A7. It has done so by comparing measurements of the Cortex-A53 cores inside the Exynos 5433 used in the Note 4 with the Cortex-A7 cores inside the Exynos 5430 used in the Galaxy Alpha. Both SoCs are produced using a similar 20nm process at Samsung, making a direct comparison possible.

Cortex-A7 is an in-order pipeline CPU core with moderate performance but an extremely small die size and very low power consumption. The Cortex-A53 core has been designed by ARM as a logical extension of Cortex-A7 to ARM's 64-bit ARMv8 instruction set with higher performance. However, in doing so die size and power efficiency have suffered somewhat.

CPU performance increased in Cortex-A53

According to the designer of Cortex-A53 at ARM, Cortex-A53 increases SPECint-2000 performance from 0.35 SPEC/MHz to 0.50 SPEC/MHz when compared to the Cortex-A7 core. In Geekbench integer benchmarks, disregarding cryptography benchmarks which a show a large increase, performance is still about 50% higher for Cortex-A53 when compared to Cortex-A7 at the same clock speed, with the biggest gains coming with multi-threaded performance (aided by the increased memory performance).

For floating point benchmarks the performance increase reported by AnandTech is dramatic, with most benchmarks showing a two to three times performance increase. However, there seems to be a discrepancy between these benchmarks results and benchmark results available from the Geekbench results database for Cortex-A53 and Cortex-A7-based devices, showing ony a moderate floating point performance increase for Cortex-A53 over Cortex-A7. Most likely, AnandTech is erroneously reporting Cortex-A57 core floating performance in this case (this matches Geekbench results that I previously tabulated).

Memory performance benchmarks performed by AnandTech show a relative increase in latency for a Cortex-A53 cluster between transfer sizes of 256 KB and 512 KB when compared to a Cortex-A7 cluster, despite the fact that this should fit inside the 512 KB L2 cache. However, as I previously noted in earlier blog articles, the benchmarks show that memory bandwidth has significantly increased with Cortex-A53 when compared to Cortex-A7, virtually doubling. This most likely contributes to the Cortex-A53 core's greater multi-threading performance in practice.

Power consumption of Cortex-A7 greatly reduced with Samsung's 20 nm process

AnandTech has published a detailed chart showing estimates for power consumption of the previous generation 32-bit Cortex-A7 and Cortex-A15 cores on both 20 nm and 28 nm processes at Samsung, based on Samsung's Exynos 5422 (28 nm) and Exynos 5430 (20 nm) SoCs.

While the high-performance Cortex-A15 cores are seeing a power reduction of about 25%, power consumption of the Cortex-A7 cores sees a significant 40% reduction with a 56% reduction at the highest CPU frequency of 1300 MHz. This can be partly explained by Samsung optimizing the Cortex-A7 cores inside Exynos 5430 for low power consumption using ARM's POP IP optimization platform.

Ironically, the excellent power characteristics of the Cortex-A7 at the latest processes such as Samsung's 20 nm process have not been taken advantage of in the market except in Samsung's Exynos big.LITTLE 5430, since Cortex-A7 adoption is mostly limited to 40 and 28 nm and all announced 20 nm SoCs use Cortex-A57 and Cortex-A53 cores. There seems to be an opportunity for ultra-efficient 20 nm Cortex-A7-based SoCs for certain product segments, while there is also a significant opportunity for 20 nm Cortex-A53-only SoCs that should be more power efficient than their 28 nm equivalents.

One could envision a hypothetical octa-core Cortex-A7-based SoC manufactured on TSMC's 20nm HPM process delivering spectacular performance/Watt, with relatively high clock speeds being possible. AnandTech's article notes that TSMC's 28nm and 20 nm HPM processes are most likely significantly more efficient than Samsung's equivalent process technology because they allow CPUs to operate at lower voltage level. A similar argument applies to Cortex-A53-based SoCs manufactured at 20 nm, albeit with lower performance/Watt.

In terms of die size, AnandTech reports a significant reduction of 45% for the the Cortex-A7 cores and 64% for the Cortex-A15 cores in the 20 nm Exynos 5430 vs 28 nm Exynos 5422.

Cortex-A53 has significantly greater power consumption than Cortex-A7

AnandTech has published a detailed chart with power consumption characteristics of the Cortex-A53 cores inside Samsung's Exynos 5433 manufactured at 20nm. In their analysis, AnandTech notes a relatively large increase in power consumption when utilizing multiple Cortex-A53 cores at their highest frequency (1300 MHz on Exynos 5433), when compared to running at 1.0 GHz. This correlates with a voltage bump when going from 1.0 to 1.3 GHz.

Based on this analysis, the article concludes the power consumption is more than twice as large for Cortex-A53 when compared to Cortex-A7 at an equivalent clock speed of 1300 MHz at a similar manufacturing process (Samsung's 20nm process). Although the Cortex-A53 core's CPU performance is greater, it is not twice as great leading to clearly lower performance/Watt for Cortex-A53 when compared to Cortex-A7.

It is possible that the chip errata (hardware bugs) in earlier revisions of Cortex-A53 that I mentioned in previous articles play a role in reducing the measured performance and power efficiency of Cortex-A53. Exynos 5433 uses Cortex-A53r0p1, which is affected by this. The chip errata require more frequent cache flushing as a work-around, which can potentially affect performance as well as power consumption. The non-optimal state of big.LITTLE kernel scheduling code may exacerbate these problems. There is potential for later revisions of Cortex-A53 such as r0p3 to deliver higher efficiency because they are not affected by these hardware problems. Chips with Cortex-A53 revision r0p3 have not yet appeared on the market.

Chip-specific core optimizations makes comparisons more difficult

It should be noted that specific optimization of the processor cores for a particular higher clock frequency target (e.g. in chip like MediaTek's MT6752 and MT6795) or low power consumption at lower clock frequency (for example, in a big.LITTLE configuration), using ARM's POP core hardening technology, has the potential skew the comparison between different chips. MediaTek's MT6752 has already been reported to have acceptable power consumption while running at relatively high maximum clock frequency, which would otherwise be incompatible with the steep rise in power consumption for clock speeds above 1.2 GHz observed in the charts for the Samsung chips.

Die size of Cortex-A53 increased compared to Cortex-A7

The die size of Cortex-A53 cores when compared to Cortex-A7 in Samsung's chips is about 1.75 times greater according to AnandTech, although it is still below one square millimeter, which is still low for a CPU. When looking at the total cluster size, which includes the L2 cache (the same amount of 512 KB for Cortex-A53 and Cortex-A7), the die size of the cluster is 1.38 times greater. The larger die size has consequences for cost-sensitive SoCs for low-end mobile devices and IoT applications, for which Cortex-A7 remains more attractive. Cortex-A7 can also be employed as an embedded CPU in a functional block such as a baseband processor,  just like Cortex-A5 is frequently used.

Consequences for mobile SoCs

The higher performance of Cortex-A53 when compared to Cortex-A7, especially memory bandwidth, makes high-clocked multi-core Cortex-A53-based SoCs suitable for mid-range performance segments. Examples of this are MediaTek's MT6752 and Qualcomm's Snapdragon 615 SoC. These SoCs also have higher GPU performance than that traditionally associated with Cortex-A7-based SoCs.

The increased power consumption and die size of Cortex-A53 causes Cortex-A7 to remain relevant, because it still delivers superior power efficiency, cost and die size, and consequently performance/Watt and performance/dollar are better than Cortex-A53. Hypothetically, a 20nm octa-core Cortex-A7 based SoC would deliver excellent power efficiency with quite acceptable performance due to higher clock speeds, and their may be a market for such a solution for smartphones. The main drawback would be that OS ecosystems such as Android are moving towards 64-bit implementations and can also make use of new cryptography instructions in ARMv8.

Sources: AnandTech (technical Exynos Galaxy 4 Note review)

Updated 1 March 2015 (Add section about core-hardening).

Qualcomm and MediaTek see challenges in smartphone SoC market

Both Qualcomm and MediaTek recently reported financial results for the fourth quarter of Q4 2014 and made projections for future periods. Both companies are seeing challenges that are already affecting their revenues and market share now or later in 2015.

Qualcomm lowers forecast for 2015 due to weakness at major customer

In their financial report for Q4 2014, Qualcomm lowered their outlook for 2015, citing as one of the reasons reduced demand from a major customer as that customer has not selected the Snapdragon 810 processor for an upcoming flagship product. This is widely believed to refer to Samsung's upcoming Galaxy S6. In fact the trend of increasing use of in-house Exynos processors already started last year, as models such as Galaxy Alpha, Galaxy S5 Mini and Galaxy Note 4 already saw increasing use of Samsung's own Exynos processors, including modem technology in some cases.

Qualcomm also mentions a share shift among major OEMs that will result in relatively more modem chips as opposed to SoCs (clearly referring to Apple, which only uses Qualcom's modem chips), as well as heightened competition in China. Recently, Qualcomm also recently announced a resolution of the anti-trust investigation by authorities in China, which amounts to a reduction in the patent royalty rate it charges to customers in China.

Qualcomm's total market share currently still strong

At the moment, Qualcomm's market share for smartphone SoCs is still strong as shown by unit shipments and revenues for Q4 2014 and Qualcomm's estimates for Q1 2015, although its product mix has shifted to lower-end products. In comparison to competitor MediaTek, Qualcomm is doing much better in terms of maintaining or growing unit shipments (with Qualcomm in fact seeing a 14% increase in unit shipments in Q4 2014), suggesting that Qualcomm is taking market share from MediaTek as products such as Snapdragon 410 and the new Snapdragon 210 take over large parts of the low-end cost-sensitive market (especially in China) where MediaTek's 3G solutions where previously dominant.

MediaTek losing market share despite successful new products

Meanwhile, although MediaTek has seen widespread adoption of its new MT6752 and MT6732 SoCs with integrated LTE modem for the cost-sensitive mid-range market, the company saw lower unit shipments in Q4 2014 and predicts a 10 to 18% revenue decline for Q1 2015, suggesting its smartphone SoC shipments are under pressure. Given the fact that the new 4G chips have higher selling prices than existing 3G chips, the revenue decline probably reflects a relatively dramatic decline in shipments of existing 3G solutions, with resulting loss of total market share, although price reductions may also play a role. MediaTek has been affected especially by the late introduction of integrated 4G solutions and the lack of a low-end 4G solution and to a lesser extend the delayed introduction of the high-end MT6795.

Captive mobile SoC use becoming more important

Within the total smartphone SoC market (and also in the tablet maket), captive supply (whereby a smartphone manufacturer uses its own SoCs in its smartphone models) is becoming more important, which affects the market opportunity for companies such as Qualcomm and MediaTek. I already mentioned Samsung's increasing use of Exynos processors, which has a significant impact as Samsung is one of the two largest smartphone manufacturers. A major Chinese manufacturr, Huawei, is also increasingly using SoCs from its own HiSilicon division, also extending to lower end models. Apple's gains in market share also has an effect (especially on the high-end market) since it uses proprietary SoCs.

In the tablet market, the low-end and Chinese white-box market is seeing a sharp reduction in shipments in Q1 2015, with market share shifting to brand names (where captive solutions are more important, such as at Samsung) as total shipments are estimated to decline dramatically. This greatly affects traditional players in the tablet SoC market such as Rockchip, Allwinner and MediaTek. Intel's strategy of subsidizing tablet SoCs has also had an impact. According to DigiTimes, the total tablet market will decline 30% sequentially in Q1 2015, with estimates of a decline of 12% for the whole year 2015.

Sources: DigiTimes (tablet market article), DigiTimes (MediaTek results), Qualcomm, MediaTek

Tuesday, February 3, 2015

Smartphone SoC market share in China in Q1 2015

Recently, DigiTimes published an article that forecasts smartphone application processor shipments in the China market for Q1 2015. Overall shipments are expected to decline by 10% sequentially when compared to Q4 2014.

The article states that Qualcomm and Spreadtrum will see the biggest declines, with Spreadtrum shipments declining almost 20% because of a lack of LTE solutions and weakness in emerging market demand.

Qualcomm sees declining shipments, but comes off a strong base

The article mentions that Qualcomm's Snapdragon 400 and Snapdragon 600 series are facing a strong challenge from comparable parts from MediaTek, with the cost structure of presumably Snapdragon 615 being less competitive than that of (presumably) MediaTek's MT6752. In earlier posts, I already compared these SoCs in some detail, and there are reasons to expect that such a cost-structure advantage for MediaTek does exist.

However, any decline for Qualcomm will come off a strong base in Q4 2014, which saw its dominance of the integrated LTE SoC market translate into market share gains (mainly at the expense of MediaTek), also in China. For the cost-sensitive part of the LTE market (covered, for example, by Snapdragon 410 and the upcoming Snapdragon 210), it does not yet face a challenge.

MediaTek struggling to reverse trend of market share decline, but product mix improves

MediaTek's market share has been negatively affected in the last part of 2014 because it was relatively late with its integrated LTE solutions, especially for the cost-sensitive part of the market (indeed, the MT6735, intended to cover the cost-sensitive segment, will only ship in Q2 2015). The article says MediaTek's shipments will decline about 15%, which suggests further market share loss in China for MediaTek after what looks like a weak Q4 2014. However, MediaTek's product mix is changing dramatically as a strong ramp of higher-end LTE SoCs such as MT6752 is offsetting declining shipments of MediaTek's existing 3G solutions which have dominated the Chinese smartphone processor market for some time.

The competitiveness of MediaTek's LTE solutions such as MT6752 gives some reason for optimism for its prospects for the rest of 2015 as these new chips ramp. However, another key product, the high-end MT6795, has been reported to have had production issues that seem to have delayed its introduction. Contrary to my original expectations, reports based on a leaked MediaTek roadmap suggest the MT6795 contains Cortex-A57 and Cortex-A53 cores in a big.LITTLE configuration, rather than a high-clocked octa-core Cortex-A53 CPU configuration, putting it squarely in the high-end segment and reducing the potential for MediaTek's chip to be significantly more cost-competitive than competitor's high-end solutions. However, no official information about the exact CPU configuration of MT6795 as well as up-to-date information about its production schedule has yet been given by MediaTek.

MediaTek's revenues for January 2015 came in at NT$17.5 billion, a level similar to the previous month. Although a year-over-year increase of 36% is apparent, the January 2014 revenue figure does not include MStar revenues. MediaTek has merged with MStar and started recognizing revenues from it in February 2014, so the year over year comparison is not very meaningful. According to DigiTimes, market watchers expected MediaTek's Q1 2015 revenues to decline no more than 10% over Q4 2014 with a strong ramp of new 4G smartphone SoCs. However, this turned out to be too optimistic.

HiSilicon gains as Huawei increasingly adopts in-house chips

Given the mentioned declines in shipments, there must be gainers to make up the balance. The article mentions strong shipments by Huawei's HiSilicon division for Huawei smartphone models. This probably reflects a trend more utilization of its own (HiSilicon) SoCs, also for more cost-senstive segments (for example, the recently announced Kirin 620 is likely to come into play).

Sources: DigiTimes Research (China smartphone AP shipment forecast for Q1 2015), DigiTimes Research (article on delay of Snapdragon 810 and MT6795)

Updated February 17, 2015.