Ryzen beats Jewtel bios update

File: 1488335937818.webm (3MB, 700x393px) Image search: [Google]

3MB, 700x393px

>>59823429
>no benchmarks
>le angry hipster youtber reviewer guy
>12min video
How will Intel recover?

File: 1491711433851.jpg (423KB, 1920x1080px) Image search: [Google]

423KB, 1920x1080px

>>59823429
Ryzen 7 is not for gayming they said.

>>59823449
That hentai looks hot, sauce?

>>59823487
This pic is a good way to spot underage and pajeets. Having ram OCed that high is neither cheap or practical.
>>59823516
Lurk moar.

File: 40eb18370a8f71ed7eb531d8130b98a0.jpg (336KB, 800x650px) Image search: [Google]

336KB, 800x650px

>>59823429
kill yourself

>>59823449
wut movie?

Looks like the same style as mezzo forte.

>>59823520
i mean that 2133MHz ram is not that far behind in terms of performance

File: 1444303840842.jpg (4KB, 223x226px) Image search: [Google]

4KB, 223x226px

>>59823487
>1700X at 4GHz and 2666 MHz RAM is better than 7700k at 5GHz and 3200 MHz RAM

Intel is fucking complete garbage trash.

>>59823520
t. butthurt schlomo

>>59823520
>Having ram OCed that high is neither cheap or practical.

Even at 2133 the 1700x got better minimum and average compare to i7 at 3200

This is a great achievement for AMD.

>>59823589
>>59823609
>tests uses a 1070
>could by a 1080 and 7700k for the cost of 1700x and 1070

As I said. That pic is great to spot the underage and pajeets.

>>59823640

Wait, does 1080 make any different in the test at 1080p?
I'm asking serious question here.

>>59823683
Have you never seen a benchmark or are you being retarded on purpose?

>>59823640
cheapest 1080 490
cheapest 1070 350
a 7700k = 350
a 1700x = 400 (can get the 1700 for 320 and have the exact same processor but that doesn't fit your narrative)

so 140$ is the price difference between gpus
50 is the difference between the cpus

and you are saying THAT is the argument of underage pajeets
Then what the fuck are you who cant even basic first grade math?

>>59823698

No, i'm seriously asking a legit question.
This image seem like the CPU bottlenecking the game at minimum with I7. >>59823487
I'm asking if i7 will got better minimum and average compare to 1700x if they all tested with 1080

>>59823520
>Having ram OCed that high is neither cheap or practical

Bullshit. 3200MHz kits are barely any more expensive than baseline 2133MHz ones and offer a solid performance increase on both Intel and AMD setups. Anybody buying 2133MHz sticks is just crippling their system to save $20.

>>59823683
If you look closely, you'll see most of the bars look pretty similar, despite using the similar CPU's with different RAM setups. That's indicative of a GPU bottleneck. What the Anon is proposing is that a cheaper CPU (7700k) combined with a more expensive GPU (1080) would give you more bang for the buck (higher average FPS with less of a GPU bottleneck).

Personally I feel that at these framerates it doesn't really matter what you go with when putting down this much money, atleast not for this particular game.

>>59823487
i7 is a bottlenecking Inturd piece of shit.

>>59823732

Yeah, i pretty notice the highest framerate must be happen either from GPU.
But my question is why i7 got the lowest minimum in all test.
Isn't Cyrsis 3 an old title that doesn't really use many cores?

>>59823708
Nigger fuck you. A 1070 with a decent cooler is $399. a 1080 is $500

7700k onsale is $300 all day. A sale just ended in fact for that price.

>>59823716
Yes the 1080 performs much better...
>>59823721
>He's underage and doesn't know how ddr4 works
What a surprise.

>>59823759
>He's underage and doesn't know how graphs work

Intel = stuttering housefire abortion.

>>59823759
>Yes the 1080 performs much better...
That still doesn't answer my question about why i7 got worst average and minimum.

>>59823791
>quick redirect
Cute, kiddo.

>>59823801

Not a redirect, just a reminder of reality.

>>59823797
Some bullshit youtber benchmark. You seriously think it's going to be done right? I already pointed out the test used a 1070 instead of a 1080.

Those aren't minimum fps either.
>>59823811
reply for you

>>59823838

So which Cyrsis benchmark video or link that is done right according to you?
Also, can you link the video from that Crysis image, I can't seem to find them from OP link.

>>59823870
>Crysis
I literately could not give a fuck. Do your own research.

>>59823881
> Do your own research.
I'm not an avid /g/ poster, i'm just a crossboarder from /ck/.
Sorry if this too much for you.

>>59823754
Well the minimum framerates in these tests is a calculated average of the bottom 0.10% frametimes generated during the test. Considering the FPS didn't drop below 59 FPS, it appears some very mild micro stutter happened, likely as a result of some background process doing things while the test was running. If the game can utilize all 8 threads offered by the CPU, this background process' interference would cause the game to suddenly not have all the resources it had otherwise, causing delayed frames, whereas a 16 thread CPU like the 1700x would have enough threads to feed the game and deal with background processes at the same time.

I'd be interested in seeing more tests focused around a 'typical' use case scenario to see how the CPU's hold up in practice. Background services, multi-tasking, streaming, stuff like that is a natural part of what you have on a normal desktop, but don't show up during benchmark comparisons that typically run clean systems, even though clean systems are rarely used in practice.

>>59823900
how do I cook a naruto ramen?

>>59823520
>3200 ram is $5 more than 2133 model
>just a bios setting away to get it there.

found the underage kid.

>>59823900
It is because I'm not searching through the 500 AMD shill videos to find the one with the retard using the 1070.

>>59823640
>white american education

>>59823926

Well you can point me one with 1080 and 700k Cyrsis 3 benchmark that was not made by some amateur youtuber.
I just want to see if it actually got better minimum compare to AMD one.
You sound like an expert compare to me.

>>59823940
>700k

I meant 7700k

>>59823940
google.

File: not-sure-if-i-can-post-this_webm_2919735.webm (1015KB, 500x400px) Image search: [Google]

1015KB, 500x400px

>>59823516
Just save it and ask on >>>/r/
Doesn't look like you're getting help here

>>59823961

Can you just help fellow anon here.
All google result that I find looks like a cluster fuck and I don't know which one that you not consider as amateur reviewer

>>59823429
why would updating bios lead to better performance?

Does it count for all performance or just specific workloads?

Does it apply to even older intel/amd cpus?
Does it apply to ram/gpus?

or just if the motherboard was awful to begin with somehow?

I am pretty dumb when it comes to hardware

>>59824005
They're all fucked. Drivers change and 1080 in one set up doesn't mean it'll get the same permanence in another. The benchmarks just give you an idea.

GPUs are a bit more difficult to skew results with in modern games though. Unless the guy is deliberately underclocking or gimping the GPU it should be alrightish.

>>59824030
>They're all fucked.
welp. I don't know which one to trust anymore

Also I just found a video that used 7700k vs 1800x using Titan X pascal GPU.
I saw that the 7700k at some point also got high drop of minimum compare to 1800x and i7 6900K.
Is the Titan X better than GTX 1070?
I can't find anything with GTX 1080

>>59824019
It would mean the motherboard manufacturers didn't have time to prepare their hardware for the CPUs properly, because AMD didn't have samples of the actual products for them to test against.

This is important, because Ryzen was an all new architecture from AMD, and learning how to get the most out of the architecture takes time. Since they didn't have this prepared before the launch, it only makes sense to see the updates come rolling in the weeks after.

You'll likely not see the same thing happen with Ryzen+.

>>59824074
I suppose intel architectures (kaby, skylake, ivy bridge) are not different enough to give such huge results?

(or does intel tend to provide samples to a larger extent?)

>>59824070
There's different versions of the Titan. That's different shitshow.

>>59824106

That still doesn't answer my question, is the Titan X pascal better than 1070?
Also the video I mention was from this one https://youtu.be/HZPr-gNWdvI?t=8m28s
I don't know if Digitalfoundery count as amateur or professional reviewer.
Just found them using google as you said.

>>59823516
Kite.

>>59823542
Yes, same director (Yasuomi Umetsu, a master).

>>59824126
titan x is better than 1070 and 1080
and digitalfoundry is solid.

>>59824165
So is that meant using better GPU still doesn't fix the 7700k problem?
That's weird considering old tittle like Crysis 3 must be using less cores and it shouldn't hinder the i7 performance at all.

>>59824192
well, there is a trend that people are saying ryzen cpus are giving much more smooth experience. minimum and maximum fps are closer to each other with ryzen than i7s.
but i7 7700k, is giving much more maximum fps and in some cases much more lower fps than ryzen counterpart
I hope that answers your question.

>>59824126
The Titan X (pascal) that launched last year should be faster, yes.

The Titan Xp that launched last week or so should also be a fair bit faster.

>>59824238

Yeah, I notice that most of 7700k benchmark got at least 5%-10% higher fps compare to 6900k/1800x, maybe because it got better cpu speed.
Still, before this I never seen any higher grade intel CPU like 6900k used as comparison test before.
Seems like more cores actually give better result, even when you test on old game at 1080p.

>>59824290
>Still, before this I never seen any higher grade intel CPU like 6900k used as comparison test before.
because there wasn't a cpu that gave 6900k performance and cost less than half of it.
>Seems like more cores actually give better result, even when you test on old game at 1080p
nope, in older games powerful fewer cores are better as we can see in the benchmarks (thanks to incompetent developers)
and kabylake arch has the most powerful single core perf. ever.

>>59823721
>3200MHz kits are barely any more expensive than baseline 2133MHz
2x8GB
2133MHz 120€
3200MHz 180€

Yeah, thanks. These 3200MHz still don't work with any AM4 motherboard anyway. There are exactly 4 models that do, some of them already not available/sold out.

>>59824340
>in older games powerful fewer cores are better as we can see in the benchmarks
That's why I'm wondering what happen to these Crysis 3 benchmarks.
The video I posted using TitanX pascal also showed that 7700k got dropped in minimum
Meanwhile the 6900k one that got more cores and less CPU speed is more stable, the 1800x also looks stable.

Is Cyrsis 3 considered as modern game or old game that benefit in better single core or not?
The game came out at 2013 where less cores CPU was still the king.

>>59824399
crysis 3 is an old dx11 game. and there could outliers. don't dwell on crysis that much.

>>59823429
>not knowing about window's
If (!genuine_intel)
sumulatebottleneck();

linux vulkan or enjoy your simulated bottleneck

>>59824399
Crysis 3 started the trend with more cores over few faster ones. Before that i3 and i5 s were awesome for gaming and i7 were for other purposes mainly. Now they moved to more threads cause of the xbone and ps4 having about 7 usable cores. So u want to game before and up to 2014ish just get any decent i3 i5. After that the hur durr i5 for gaming ,save the 100 dollars is a meme for retards. i built my bro a 4690k at 4.7ghz and 2400mhz ram,980 machine and is a stuttering piece of shit. Dropped a stock clocked 4770 non k and was silky smooth. i even got it to 4 3 ghz with turbo and bclk.

>>59823759
Ashy skin of the Singularity

>>59824091
This and compatibility improvements in general have been the case with Nehalem and Sandy Bridge too. Both of them were also buggy as hell at the beginning because they were very different architectures than the previous generation.
That's especially true for construction cores -> Zen in that it's one of the most radical architectural changes we've seen in x86 processors for many years.
On top of that, mainboard manufacturers were doubting Zen's success, and didn't start working on Ryzen mainboards until very late in its development cycle. That's mostly the reason for Ryzen's bad RAM compatibility and part of the reason for the limited mainboard availability right now.

File: good goy.png (604KB, 800x523px) Image search: [Google]

604KB, 800x523px

>>59824638
It's like everyone (((convinently))) forgot what a new architecture launch looks like.

>>59825005
I still can't believe that was a thing.

>>59825005
Please delet dis

>>59823640
or just get a 1700 and overclock it since its all the same cpu

>>59823449
Did he die?