LightBlog

jeudi 8 février 2018

Ryzen 3 1300X Review: A Budget Friendly 4 Core CPU from AMD!

Given that we have tested almost the rest of the AMD Ryzen lineup, including now the Threadripper 1950X, we wanted to see what things look like in Ryzen 3. So for that we reached out to AMD and received a Ryzen 3 1300X for testing. Currently listed on Newegg at $124.99, the processor packs 4 cores/threads that seems unbeatable at that price. Can developers on a tight budget get their builds accomplished? Just how well does it perform? Is it worth someone currently on an older platform to consider the upgrade? Let’s run some tests on it and see what we can find out.

Editor’s Note: The Ryzen 3 1300x was provided to XDA by AMD for review purposes.

Unboxing, Installation & Test Setup

The Ryzen 1300X is an AM4 socket, but at its price point its audience isn’t the same as a Ryzen 5 or 7. It’s more likely that a consumer purchasing this will be likely to do an upgrade to an existing system, so the test setup this time will reflect that. We’re using the shell of an Acer Aspire M5621, so this case only supports a mATX motherboard. At the moment options in this space are not as generous as the ATX offerings, but MSI’s B350M Bazooka stays within a reasonable price of $89.99 (on Amazon) and continues the narrative that the PC is a piecemeal upgrade.

To avoid potential bottlenecks elsewhere, we will use a higher end graphics card, DDR4-2400 RAM, and m.2 SSD. This should ensure that the test results are limited by CPU only. And as usual, items not purchased by either myself or XDA will be noted as to who provided it to us.

Platform Configuration

Software/Operating System

  • Ubuntu 17.10
  • Phoronix Test Suite—current using apt

Testing Methodology

Given our most recent batch of complete tests were based on stock speeds, we can now include these two CPUs in the batch by keeping at stock as well. A brief test showed that the Ryzen 1300X could be overclocked easily to 3.8 by multiplier only, but 3.9 was not possible. But we’re focusing this review specifically between the 1300X and 1500X because in many respects they are identical. The main difference between the two is the addition of simultaneous multi-threading (SMT) in the 1500X. Both have the same stock speeds and will offer us a great chance to see how much of a difference SMT makes in performance.

Non-Build Benchmarks

Benchmark Notes: Phoronix Test Suite’s CPU suite offers a plethora of tests and not all will be included in this review. The full test results for the tests from the Phoronix Test Suite are available on OpenBenchmarking.org.  Benchmarks are color grouped by Ryzen/Threadripper, Intel 7700K/8700K and then Intel high end desktop CPUs separately in a color scheme that more closely follows other graphs on XDA.[/alert]

FFTW

FFTW is a single-threaded benchmark of Fast Fourier transform. In comparison to each other there isn’t much of a difference, and we would expect that based on single-threaded performance. 

GZip Compression

GZip is a common compression method and so it makes sense to check out the performance here. Since both should perform identically in regards to speed, we have to suspect that SMT did actually play a slight role here in the 1500X decrease.

SciMark 2 (Java) v1.3.0

The SciMark 2 benchmark utilizes Java for arithmetic operations and then provides scoring based on those results. Here we see these floating within the same range as the 1700X at stock, which also is unsurprising as a single threaded test.

John The Ripper

On the cryptography front, John The Ripper is a bit surprising. The 1500X performs at nearly half of what the 1700X did during our HEDT CPU review. What’s noteworthy is that this isn’t the same case when removing SMT from the equation. The decrease is closer to a third in this case.

C-Ray v1.1

The C-Ray results from this group follows a similar trend from 1700X to 1500X—a nearly 50% reduction in performance. But the difference between the 1300X and 1500X is even less than John The Ripper. Here we’re looking at a decrease of less than 10% without SMT. Quite impressive!


Benchmarks: Build Performance

Build Test: ImageMagick

Build tests really start to muddle the scaling with core and thread counts in Ryzen. For reference, our 1700X performed this test most recently at 44.03 seconds. With half the cores and threads, we’d expect this to double, but instead it only increases by 50%. 4 cores alone nearly doubles the build time from the 1700X and suggests in builds the higher core and thread counts really do make more of a difference at this level.

Build Test: GCC

We find similar results here. Compared to the 1700X results of 902.06, the 1500X is approximately 50% longer and the 1300X nearly doubles it again.

Build Test: LineageOS cm-14.1 Pixel XL

We stay with the Pixel XL for timed build benchmarks. As requested by readers, the graph will display both build times with and without caches. But the results without cache follow a similar trend to the other two build tests.

Conclusion

So, what does a 1300X get you? For nearly a third of the cost you’ll get about half the performance in builds. Single-threaded performance will largely be the same and vary based on clock speeds more than anything else. And once you move into cached builds, getting a cached build within 35 minutes isn’t bad at all. You’ll want to take a look at what your use cases are beyond the builds and see if you’ll put those additional cores and threads to use. Even if you decide to go a little higher, the 1500X offers about 75% of the performance on building for about half the price.

All in all as we look at the entire Ryzen lineup, we see they have indeed come a long way. AMD has made great strides in really offering great performance at nearly every price level, and the 1300X is no exception to that. It will be even more interesting to see how these two in particular fare against their upcoming counterparts with RX Vega graphics integrated into the CPU. We also look forward to their second generation lineup that’s supposed to roll out later this year. But for a first year redesign? Great job AMD—at all levels!



from xda-developers http://ift.tt/2EcpJCa
via IFTTT

Aucun commentaire:

Enregistrer un commentaire