AMD Ryzen 5 1400 review
Click here to post a comment for AMD Ryzen 5 1400 review on our message forum
Herem
Nice review and a good budget chip, but I agree with the summary that the age of quad core processors is drawing to an end.
@Hilbert is there any chance you will be doing a more in depth article about memory speeds on AGESA 1.0.0.6? It would be great to see how far the speeds can be pushed using a few different configurations of both single and double DDR4 DIMMs in both 2 and 4 stick setups.
kapu
Holy crap , it matches 6600K .. Really nice performance.
But we need results of i5 7600k and i5 7500, why no such results Hilbert ?
i5 7500 is priced 20 euro more in Poland and it is direct competition.
jdc2389
Please add the gaming benchmark results at the overclocked 3.9 Hilbert! Thanks for the review.
AndyMutz
was i lucky with my chip?
mine 1400 does 3900 MHz at 1,33v on a gigabyte GA-AB350-Gaming 3.
4000 MHz ran unstable at 1,35v and i did not want to go higher with the voltage.
-andy-
Guru3dreader
CPU-Z benchmark is a joke. I wouldn't give it even a line in a review, not a whole page. Are we going back to the times where people where using SuperPi to check performance? Especially those times when slower Intel processors where finishing it much faster?
Fox2232
I think we can successfully remove CPU-z bench from set. I just updated to 1.79.1.x64.
And its results are not compatible with 1.79.x64 again.
And the Fun(ck) part is: It puts my i5-2500k @ 4,5GHz multi-threaded score 11% above FX-8350 @ 4,0GHz and that would mean double IPC.
But FX-8350 would win in any sane multi-threaded benchmark. In geekbench it does like 50% better without crazy OC.
Simply, for me CPU-z bench has no longer any informative value.
MainFrame Alpha
thank you Hilbert for yet another great review, this is a great value product indeed for for the budget segment level 🙂
and thanks for the added DDR4 chart, i think it is most helpful for those in need for a quick guide in which ram to pair with Ryzen 🤓
Fox2232
Watching Ryzen 7 1800X w/ 3600MHz memory results in:
AMD Ryzen 5 1400 review - Performance - System Memory Perf: DDR4 Memory Read / Write
:) That's quite a step up from 3200MHz.
BlueRay
"With the Ryzen 5 1400 you enter the mainstream gaming segment, paired with a GTX 1060 or Rx 570/580 you will not run into CPU bound limitations, hence such a setup would be a perfect symbiosis."
So my question is what happens when you pair any Ryzen with a faster GPU? Nobody is buying a CPU for only 1 year. Faster cards will come very soon and assuming someone will keep it for 4-5 years this can spell problems?
Hilbert Hagedoorn
Administrator
Within that mindset you could not buy a similar priced Core i5 7400 either right? But sure, 4-5 years from now if you still are at 1080p gaming yet have an Ultra fast graphics card you run into CPU bottle-necking (which has been around for years and is the same with Intel mainstream SKUs).
CPU bottlenecks are nothing new, and in 5 years from now your games will be more graphically demanding as well balancing out that extra graphics horsepower as well. A GPU bottleneck far is more important on its influence on framerate then a CPU bottleneck.
vipin
L3 cache shown as 8MB in page 1
@Hilbert since Ryzen 5 1400 has 2 dedicated cores from each CCX should not overall L3 cache be shown as 16MB or 8MB/CCX? Did I miss something?
Italia1
An italian site has done almost same test (Bit and chips). This cpus look to be very good
Amx85
Good review, but i still interested in non-X R5-1600 CPU, in some days i'll upgrade to 4GB RX460 or 2GB RX560 since i got an AOC 1080p LED monitor and my R7-260X + FX-8320E does'nt get +40fps in some games at that resolution.
You forgot update ROTR? last patch has CPU performance optimizations, specially for RYZEN
Thanks and good morning
Turanis
Not bad vs Intel i5 Skylake (Kaby Lake is an OC Skylake Cpu).
AndyMutz
the 1400 has 4MB/CCX, so 8MB total.
-andy-
Loophole35
All hail the new budget gaming KING. No point in the i5 anymore.
Fox2232
HH tests at stock clocks. If you get Ryzen 5 1600, you will have no trouble clocking it at least to 1600X stock clock. So you can use that as reference.
schmidtbag
Those were some pretty high OC voltages despite not reaching 4GHz.
I personally would've bought the 1400 if it was just a single CCX, because I think it'd have been noticeably faster and probably more capable of higher memory clocks. The smaller L3 doesn't bother me; 16MB was intended to be enough for the 8-core models, so surely 8MB is plenty for a 4-core. Though there are tests showing the smaller L3 does hurt performance (compared to a similarly clocked 1500X), it wasn't a big difference. Anyway, since this CPU is basically just a crippled 1500X with a worse heatsink for only $20 less, this CPU is pretty much on the borderline of not being a good value; if you honestly don't care about the extra L3 and have no intention to use the stock heatsink then I suppose you might as well save yourself the $20 and put that toward the heatsink.
I hear the Ryzen 3 models are supposed to utilize a single CCX. With SMT disabled too, it'll be very interesting to see their single-threaded performance.
Amx85
i know, and the 1600 has stock cooler :infinity: and 1600X no :bang:
waltc3
Agreed. What makes it even sillier is the fact that the guy didn't like the fact that it made the Intel cpus look bad so he rewrote it specifically to make the Ryzens appear slower in the comparison...;) Hopefully, people catch a glimmer of how benchmarks can be used--and often are--to mislead. He's certainly not alone in playing favorites with his benchmarks, though. Benchmarks are very much like election polls--rarely accurate, and exist primarily to shape and form opinions and biases.
