AMD A10-7850K Performance Review | Puget Systems
Always look at the date when you read an article. Some of the content in this article is most likely out of date, as it was written on January 6, 2014. For newer information, see our more recent articles.
Table of Contents
Introduction
Alongside the A10-7700K, the A10-7850K is a recent addition to AMD’s A-Series line of APUs and uses the existing FM2+ socket. By using the new 28nm manufacturing process, these new APUs are able to run faster than the previous generation, even when running at a lower clock frequency. In addition, improvements in the graphical portion of the APU increases the capabilities of the integrated R7 graphics controller. Interestingly, the clock speed of these new APUs are actually lower than the models they are replacing.
While new computer hardware is almost always faster than the models they are replacing, it is often hard to get through all the marketing talk to find out exactly how much better they are. For that reason, we are going to be comparing the performance of the A10-7850K to multiple CPUs and video cards. In addition, since the ability to use high frequency RAM is often cited as an important feature of the A-Series APUs, we will also be looking how much performance gain you actually will see by using high frequency RAM.
Test Setup
To perform our comparison benchmarks, we used the following hardware:
*Shared memory set to 1GB. RAM model will be adjusted according to the testing performed.
The multiple models of RAM will be used to determine how much of a performance loss or gain there is by using RAM that has a higher frequency, but looser timings. The A10-7850K natively supports up to DDR3-2133, so that is the fastest RAM we will be testing with. Note that the shared memory will be set to 1GB to keep it consistant between the Intel and AMD systems. As we showed in our Optimizing AMD Trinity for Budget Gaming article, even 512MB of shared RAM would be more than enough RAM for the settings you would use with these onboard graphics – so 1GB of shared memory should ensure that the amount of shared RAM should never be an issue.
The A10-6800K and Intel test system will be used to compare the A10-7850K against both the previous generation APUs as well as against an equivelantly priced Intel system. The two motherboards we are using are almost the exact same price and the CPUs are within $10 of each other depending on where you purchase them.
Impact of RAM Speed on CPU Performance
Starting with how the RAM speed should affect programs that are primarily CPU-based, we started with PCMark08 and ran both the Creative suite and Microsoft Office suite of tests.
As you can see, there is a difference, but it is not very much. The DDR3-1866 RAM performed within 1% of the DDR3-2133 RAM, and the DDR3-1600 RAM was within 3%.
In Cinebench R15, we actually see the DDR3-1866 RAM outperforming the DDR3-2133 RAM. Once again, however, the difference is very, very small – only about 1% – so the actual real-world difference is negligible.
Geekbench is one of the few completely synthetic benchmarks we run, but it gives some very useful theoretical performance data. Interestingly, the results are similar to the Cinebench test as the DDR3-1866 RAM is again the fastest. Just like Cinebench and PCMark, however, the difference between DDR3-1866 and DDR3-2133 is only about 1%. DDR3-1600, on the other hand, shows a bit larger of a difference, especially in the Memory score which measures the bandwidth performance of the RAM.
Impact of RAM Speed on Gaming Performance
RAM speed plays a much bigger role on gaming performance when using integrated graphics, so we expect to see a much larger performance difference than we saw with the CPU-based applications.
While we tried to use video settings that would get us to around 60 FPS when using the DDR3-2133 RAM, some titles – like Hitman: Absolution – simply could not reach that 60 FPS target even running at minimal settings. While the exact difference in performance varies by the game, using DDR3-1600 RAM resulted in an average FPS drop of about 23.5%. The DDR3-1866 RAM was better, but still resulted in about a 5% drop in performance.
We do want to throw one little caveat in here, which is that while DDR3-2133 RAM did perform the best in our testing, it is only going to do so if you are using the integrated graphics. Once you use a discrete video card, our testing has shown that the speed of the RAM makes almost no difference in games. Also, we know from experience that higher frequency RAM is more prone to failure than lower frequency RAM. So if you do not need those last couple of FPS, we would recommend using DDR3-1866 or even DDR3-1600 RAM whenever possible.
APU/CPU CPU Performance Comparison
To get an idea of how the A10-7850K compares to both the previous generation A10-6800K and the Intel Core i5-4440, we performed benchmarks on each APU/CPU to see how they compare. For these benchmarks, we will be using DDR3-2133 RAM on the AMD APUs and DDR3-1600 RAM on the Intel CPU. We decided to use those speeds of RAM as they are what the CPU is natively rated to be able to use.
The A-series APUs from AMD have never really been able to compete with Intel CPUs on CPU-based tasks, which is pretty clearly shown in the PCMark 08 benchmark. What we didn’t expect was how much better the A10-7850K performs compared to the A10-6800K. Even though the A10-7850K runs at a lower clock speed than the A10-6800K, the newer architecture allows it to be faster even at that lower clock speed.
Oddly, in CineBench R15 the A10-7850K is slower than the A10-6800K. This pretty much directly counters the PCMark 08 results, although we would consider PCMark 08 to be the better measure of normal system performance.
In Geekbench, the A10-6800K is actually faster than the A10-7850K in the single-core tests, but the A10-7850K is faster in the multi-core portion of the benchmark. This is interesting and indicates that the A10-7850K is not as good as the previous generation APUs when it comes to single-core applications. For anything that is multi-core capable, however, the A10-7850K should outperform the previous generation APUs.
One thing that is clear across all of these benchmarks is that the Intel Core i5-4440 greatly outperforms the AMD APUs in CPU-based applications. So to show the advantage of the AMD A-Series APUs, let’s next look at gaming performance using the integrated graphics.
APU/CPU Gaming Performance Comparison
With these benchmarks, we get a good look at the main strength of the A10-7850K. AMD APUs have pretty much always out-performed the Intel integrated graphics, and the A10-7850K simply increases the difference. On average, the A10-7850K is about 40% faster than the Intel Core i5-4440 and about 10% faster than the A10-6800K.
In fact, this performance bump means that the A10-7850K is actually able to match some low-end discrete graphics cards. In our testing, we found that it performs almost exactly between the AMD Radeon R7 240 and the NVIDIA Geforce GT 640.
Conclusion
Overall, the AMD A10-7850K is a great improvement over the previous generation APUs in AMD’s A-series line. Especially in terms of gaming performance when using the integrated graphics, it is about 10% faster than the previous generation APUs. If you compare the gaming performance to an Intel CPU, the graphical capability is so much better that it really is not even in the same league.
In terms of CPU performance, the A10-7850K is overall an improvement over the A10-6800K. Especially in PCMark 08, the A10-7850K was about 10% faster than the A10-6800K. Our GeekBench benchmark indicated lower single-core performance, but as more and more programs use multi-core technology, that is really not even much of an issue.
However, there are a few issues we have with the A-series line in general. First, while the gaming performance is better than Intel, it is only better if you are using integrated graphics. As soon as you add a discrete video card, an equivalently priced Intel-based system is going to give you overall better performance in most games. And while the gaming performance is very good for integrated graphics, if you consider yourself even a moderate gamer you likely would want to invest in a discrete video card.
Similarly, if you are mainly concerned about performance in programs that do not use the GPU, an Intel-based system is going to give you much better performance for your dollar. Finally, in terms of future upgrade potential, the A-series APUs hit a wall much, much sooner than their Intel equivalents. The A10-7850K is the fastest A-series APU currently available, yet the CPU performance is much lower than even the Intel Core i5-4440 which is only a mid-range Intel CPU.
All that being said, there are a number of applications in which the new AMD APUs are uniquely suited for. In particular, Media center PCs that either don’t need or have space for a discrete video card would be perfect for the A-series APUs. They have plenty of graphical power for most tasks a media center PC would be needed for, and can even cover light gaming duty when needed.
Tags: 7850K, A-Series, AMD, Kaveri
AMD A10-7800 Kaveri APU Review
Although AMD promised many new APUs when unveiling its Kaveri architecture in January, the company’s lineup has been limited to the $160 A10-7700K and $170 A10-7850K, with no retail appearance by the A8-7600 that we tested seven months ago or any other budget Kaveri chips for that matter.
It’s a little confusing to see the company pitching an updated version of a product range that was never properly launched but on the bright side, AMD’s Kaveri update seems a little more substantial than Intel’s recent Haswell refresh, if only for the fact that the A8-7600 is finally set to appear at retail.
Along with the A8-7600 due to arrive for $100, AMD is also shipping the $77 A6-7400K and $155 A10-7800, the latter of which being noteworthy because it has 45/65W TDP modes like the A8-7600 except it gets a full-blown R7 GPU featuring 8 CUs with 512 SPUs and 32 TAUs like the unlocked A10-7850K.
| A10-7850K | A10-7800 | A10-7700K | A8-7600 | A6-7400K | |
| Compute Cores | 12 (4 CPU + 8 GPU) |
12 (4 CPU + 8 GPU) |
10 (4 CPU + 6 GPU) |
10 (4 CPU + 6 GPU) |
6 (2 CPU + 4 GPU) |
| Max Turbo / CPU Frequency | 4.0/3.7GHz | 3. |
3.8/3.4GHz | 3.8/3.1GHz | 3.9/3.5GHz |
| HSA Features | Yes | Yes | Yes | Yes | Yes |
| AMD TrueAudio | Yes | Yes | Yes | Yes | Yes |
| Mantle Support | Yes | Yes | Yes | Yes | Yes |
| AMD Configurable TDP | Yes | Optimized (45w) | Yes | Optimized (45w) | Optimized (45w) |
| Suggested Configuration | Radeon R9 270, A88X, 8GB DDR3-2400 | Radeon R7 265, A88X, 8GB DDR3-2133 | Radeon R7 260X, A78, 8GB DDR3-2133 | Radeon R7 260, A78, 8GB DDR3-2133 | A58, 8GB DDR3-1866 |
| Price | $173 | $155 | $153 | $101 | $77 |
All models support DDR3-2133 memory along with the new Socket FM2+ with chipset support for PCIe 3.0. Since Kaveri brings some pretty significant changes, it’s unsurprising that a socket update was in order.
Rather than go over the Kaveri architecture again, please refer back to our article published in January. With that said, let’s jump right into testing.
|
|
||
|
The AMD A10-7800 was tested using dual-channel DDR3-2133 memory which allowed for a bandwidth of 14.2GB/s, slightly faster than what we saw from the A8-7600.
The A10-7800’s cache performance is also very similar to the A8-7600 while its L1 cache is considerably slower than the Core i3-4130’s.
Overview of AMD A10-6700 and A10-6800K processors
AMD continues to release new processors based on the old architecture. Thus, the corporation postpones the release of fresh solutions on a modern technical process. In graphic segment the situation is identical. But autumn is coming, and we have hope for the long-awaited announcements, because users want to see more productive devices on the market.
The A10-6700 and A10-6800K CPUs cannot compete with Intel when it comes to physical core performance. nine0009
The chips reviewed in this article (the «Richland» family) are a logical continuation of the AMD A10 hybrid lineup. These models are practically no different from previous 5000-series CPUs (eg A10-5800K).
Specifications
AMD A10-6700 and A10-6800K gems are compatible with Socket FM2. The first microprocessor is equipped with four physical cores (nominal frequency — 3700 MHz, L2 cache — 4 MB). The maximum supported memory standard is DDR3-1866. Surprisingly, the TDP of the A10-6700 does not exceed 65 W (not bad, considering that in mode turbo this CPU operates at 4300 MHz).
AMD A10-6700 and A10-6800K have the same graphics, HD 8670D. In the asset adapter 384 computing units , the nominal clock frequency of the core is 844 MHz.
| A6-5400K | A10-5800K | A10-6700 | A10-6800K | |
|---|---|---|---|---|
| Socket | FM2 | FM2 | FM2 | FM2 |
| Process | 32 nm | 32 nm | 32 nm | 32 nm |
| Cores/Threads | 2/2 | 4/4 | 4/4 | 4/4 |
| Clock frequency | 3600/3800 MHz | 3800/4200 MHz | 3700/4300 MHz | 4100/4400 MHz |
| L3 cache | 1 MB | 4 MB | 4 MB | 4 MB |
| TDP | 65W | 100W | 65W | 100W |
| Graphics | Radeon HD 7540D | Radeon HD 7660D | Radeon HD 8670D | Radeon HD 8670D |
| Memory support | DDR3-1866 2 channels |
DDR3-1866 2 channels |
The A10-6800K variant boasts support for faster RAM (DDR3-2133), an unlocked multiplier (don’t forget the K in the name), and an impressive clock speed of 4100/4400 MHz. TDP older brother — 100 watts.
Test stand:
Maternal board-ECS A55f2-M3
RAM-G.Skill Ripjawsx F3-12800CL10D-16GBXL
performance and results of testing
,000
Sixth series of processors AMD rose compared to AMD compared to AMD CPU of the 5000th line), but only in relation to physical cores . Integrated graphics are less impressive.
AMD A10-6700 and A10-6800K have the same graphics, HD 8670D.
The AMD A10-6700 and A10-6800K lose almost twice as much to the A10-5800K in 3D benchmarks.
It is possible that the case lies in the motherboard. On a top-end based with a high-end chipset, the situation may be completely different, although there are no guarantees (the situation is strange anyway).
AMD A10-6800K easily overclocked to 4800 MHz and even started at 5 GHz (but this is already an unstable result). nine0009
Conclusion
The A10-6700 and A10-6800K CPUs can’t compete with Intel’s (even entry-level) solutions when it comes to physical core performance. AMD developers urgently need to change something and improve the architecture. Increasing the number of threads and clock frequency will not correct the situation, and the 32 nm process technology has long exhausted its potential.
Tags: AMD
Processors amd a10 specifications — Dudom
Today we have another “pass-through” testing on our agenda, which is connected primarily with a change in the test methodology, and not with the release of new processors. Moreover, the topic will be a platform where things are not going well with the release of new models, despite all its attractive sides: it has been more than a year since the A10-7850K appeared, and it continues to be the most powerful solution in the line. Moreover, no significant change in the situation is planned in the near future. Around the middle of the year, the A10-8850K should appear, but apart from a very slight increase in clock frequencies, nothing is expected from it (as well as from other models of the updated line). Thus, it makes no sense to specifically wait for the appearance of Kaveri Refresh processors, and something more interesting may appear only next year (and within a completely new platform, most likely completely incompatible with today’s one). In general, if there is a need to purchase something, it can be done now. Moreover, it is not even necessary to choose Kaveri — although the shipment of processors based on the previous architecture has already been stopped, they are still found in the distribution network, moreover, at more attractive prices than modern models. There is only one question: is the economy reasonable? And do you need a top model? But it is impossible to find out without tests, so now we will start with them. nine0009
Test stand configuration
| Processor | AMD A10-6800K | AMD A10-7800 | AMD A10-7850K |
| Kernel name | Richland | Kaveri | Kaveri |
| Production technology | 32 nm | 28 nm | 28 nm |
| Core clock std/max, GHz | 4. |
3.5/3.9 | 3.7/4.0 |
| Number of cores (modules)/computation threads | 2/4 | 2/4 | 2/4 |
| L1 cache (total), I/D, KB | 128/64 | 192/64 | 192/64 |
| L2 cache, KB | 2×2048 | 2×2048 | 2×2048 |
| L3 cache, MiB | — | — | — |
| RAM | 2×DDR3-2133 | 2×DDR3-2133 | 2×DDR3-2133 |
| TDP, W | 100 | 65/45 | 95 |
| Graphics | Radeon HD 8670D | Radeon R7 | Radeon R7 |
| Number of HP | 384 | 512 | 512 |
| Frequency std/max, MHz | 844 | 720 | 720 |
| Price | $138(73), T-10387700 | $154(66), T-10674780 | $162(67), T-10674781 |
A10-6800K and A10-7850K at the end of last year and came to the conclusion that these models are approximately equivalent in performance, but the first one is cheaper. However, updating the software in the new version of the methodology may well lead to the fact that the alignment will change — this is what we will check. At the same time, adding the A10-7800 to the test subjects: it is a little more economical and a little slower than the top model, which is interesting. Note that the 7800 is actually the only significant expansion of the range of processors for FM2+ in 2014: earlier desktop A10s based on Kaveri did not fit into a TDP of 65 W or less. If we consider work with a heat pack reduced to 45 W (which may be relevant for a compact solution), the situation is even aggravated by the fact that AMD previously released only a couple of suitable models for FM2, which were not so easy to purchase. Now the problem is gone. And the only question is what will be the performance loss. Especially against the background of top models, which obviously “do not fit” into small Mini-ITX cases due to a “serious” heat pack. nine0009
| Processor | Intel Core i7-5500U |
| Kernel name | Broadwell |
| Production technology | 14 nm |
| Core clock std/max, GHz | 2. |
| Number of compute cores/threads | 2/4 |
| L1 cache (total), I/D, KB | 64/64 |
| L2 cache, KB | 2×256 |
| L3 cache, MiB | 4 |
| RAM | 2×DDR3-1600 |
| TDP, W | 15 |
| Graphics | HDG 5500 |
| Number of HP | 96 |
| Frequency std/max, MHz | 300/950 |
Who do these processors compare to? Such a question is always relevant at the beginning of the testing cycle — the base of already obtained test results is too small. Therefore, by a strong-willed decision, we did not look for competitors «on the forehead», but took the numbers obtained when testing the Core i7-5500U. It is clear that the model is ultrabook, though. Although many people these days are concerned about the direct comparison of laptop and desktop performance, it is interesting to look for a direct answer to it. nine0009
Test methodology
To evaluate performance, we used our performance measurement methodology using iXBT Application Benchmark 2015 and iXBT Game Benchmark 2015. We normalized all test results in the first benchmark against the results of the reference system, which this year will be the same for laptops , and for all other computers, which is designed to make it easier for readers to compare and choose:
| Processor | Intel Core i5-3317U |
| Chipset | Intel HM77 Express |
| Memory | 4 GB DDR3-1600 (dual channel) |
| Graphics subsystem | Intel HD Graphics 4000 |
| Accumulator | SSD 128 GB Crucial M4-CT128M4SSD1 |
| Operating system | Windows 8 (64-bit) |
| Intel graphics driver version | 9.18.10.3186 |
iXBT Application Benchmark 2015
Despite the fact that these applications load the processors «to the full», including the video core, the ultrabook Broadwell turned out to be not much worse than the AMD desktop APUs — in fact, the level of performance it demonstrates is A10 -6800K and A10-7800 in «45W mode» are the same. But in normal mode, the A10-7800 is noticeably faster, but the A10-7850K outperforms it quite a bit, which makes it not the best choice for today. nine0009
Here, the top AMD APUs are completely behind the Core i7-5500U, which can only be called a mockery 🙂 The distribution of places among them almost does not change. Unless the A10-6800K managed to keep up with the A10-7800 with a “squeezed” heat pack, but if this is not necessary, then the latter is preferable in new software versions.
Photo manipulation is similar to video processing. True, the performance requirements on the part of the user are lower here, since the slowness of the computer (if any) interferes less — the workflows themselves are shorter. nine0009
Adobe Illustrator cheerfully changes version numbers, but the software algorithms themselves seem to be the same as 10 years ago. With such a curious effect, the flagship for FM2 is even slightly faster than the new top solution. However, slightly.
In Audition, the A10-6800K and A10-7850K are roughly equal in favor of a newer processor. But a comparison with the i7-5500U shows that this is just another «inconvenient» case for AMD. Quite inconvenient — where desktop processor models lose not only to laptop ones, but also to ultrabook ones (and if things continue like this, they will soon start tablet ones). nine0009
In the previous version of the program, the A10-6800K kept at the level of the A10-7800 with a TDP of 65 W, but now it has slipped to parity from 45 W: as we can see, the software update has a positive effect on intracompany competition. However, the effect somehow obtained is not enough 🙂 rulers. The lag, however, is microscopic, but it exists. What is especially frustrating against the background of the fact that Intel now even CULV solutions are noticeably faster at times. nine0009
All are approximately equal, except for the A10-7800 in the reduced TDP mode — apparently, to save energy, the processor tries to spend most of its time in sleep mode, the exit from which takes a certain time, which is especially noticeable under such types of load.
And file operations are somewhat similar to, for example, archivers, which is not surprising — unpacking an ISO image is ideologically close to them. Formally, however, these subtests hardly load the processor with work, so the difference between them is mainly due to the peculiarities of the power-saving modes. nine0009
Once upon a time, older models of AMD APUs successfully competed in processor performance with desktop Core i3. Now, as we see, dual-core processors for ultrabooks are also able to overtake them, with which we “congratulate” the company. In general, the need for modernization in this family is long overdue. And it is a pity that it will not be at least another year. Yes, and anything can happen — a software update in the test method, of course, allowed the A10-7850K to bypass the older A10-6800K in the overall standings, but some 5% performance increase, it seems to us, is not at all what was required. The main effect of the release of Kaveri was the possibility of producing more economical models, such as the A10-7800. The fact that this processor performs at the level of the A10-6800K with a much “narrower” thermal package is already good. Although it is still not enough to compete with Intel in the field of processor performance. But the company’s products have such a trump card as a powerful graphics core. Let’s try to play it. nine0009
Gaming applications
For obvious reasons, for computer systems of this level, we limit ourselves to the minimum quality mode, not only in “full” resolution, but also with its reduction to 1366 × 768 (Core i7-5500U was not tested in this mode , but this is not too important for us now — for a qualitative comparison, one mode is enough). Despite the fact that the integrated graphics of the A10 line of desktop processors are the best on the market, even they are not yet able to satisfy the demanding gamer in terms of picture quality. But if you voluntarily agree to the «minimum salary», you can save a lot. We already know this well from previous tests, but today we’ll just see how our updated gaming set works on these processors. nine0009
The GPU performance of the A10 is almost twice that of the HD Graphics 5500, which is no secret. And the result is the ability to play, albeit in minimal quality, but in full Full HD resolution.
The game is very processor-dependent, and it mainly requires «single-threaded» performance, so it’s impossible to break away from Intel solutions here. But from a practical point of view, this is not so important — the main thing is that you can play everything.
As in Grid2. Where, however, the requirements for the GPU are higher, so that there is no difference between the subjects.
Metro series games are just very demanding on graphics, so here A10 is still not enough for FHD, but enough to play at lower resolutions.
In Hitman, older Kaveri are almost enough for FHD, and at a lower resolution you can play comfortably.
Thief is still too «heavy» for integrated graphics, although there is some progress in this area, so the processors of the Kaveri Refresh line may already «stretch» at least low-resolution modes. nine0009
Tomb Raider feels at ease even in the «full» resolution mode — here Intel processors are only a little «not enough». In general, it is already somehow possible to play such toys.
And two examples of how the transition from Richland to Kaveri gives not only a quantitative but also a qualitative effect in the absence of intercompany competition.
Total
Well, as you can see, the software update has had a beneficial effect on Kaveri: the new flagship is now at least faster than the old one, because earlier their equality caused difficult feelings, to put it mildly 🙂 However. However, it is the A10-7850K that still does not look very interesting, since the A10-7800 appeared, the performance of which is only slightly lower, and the cooling requirements are “lower by many”. However, this still does not pull on a radical breakthrough, since APUs are actually weaker than Intel solutions “as processors”. And even weaker than the solutions of completely different classes — in one segment it could still somehow be endured. But in different ones, the effect is poorly compensated even by a decent video core, since a gamer should still definitely look towards discrete graphics, since its use in desktop systems (even if compact ones) does not pose any difficulties. Therefore, the A10 desktop family remains a niche solution: quite expensive, but, nevertheless, not too gaming and not too productive outside of games. Moreover, cosmetic improvements are clearly unable to significantly change the state of affairs — here it is no longer necessary to rearrange the beds, but change the girls 🙂
Description
AMD started AMD A10-7850K sales on January 14, 2014 at a suggested price of $173. This is a desktop processor based on the Kaveri architecture, primarily designed for office systems. It has 4 cores and 4 threads and is manufactured using 28nm process technology, the maximum frequency is 4000, the multiplier is locked.
In terms of compatibility, this is an FM2+ socket processor with a TDP of 95W and a maximum temperature of 74°C.
