AMD Ryzen 7 1800X 3.6 GHz

[medi01]

This is the best and most informative

Oh please...
No 4k (yeah, god forbid people actually using GPUs used in the test would know how it runs on their monitors), no 8 cores from Intel, no DX11 vs 12, Ryzen using slower RAM than 7700k (after AMD said infinity fabric uses RAM clock)

"Most informative" right.

 

[notb]

Oh please...
No 4k (yeah, god forbid people actually using GPUs used in the test would know how it runs on their monitors),

You assume people actually have 4K LCDs?

no 8 cores from Intel

AMD categorized Ryzen as gaming segment. TPU is primarly a gaming-oriented site and so is the review.
In the blue camp HEDT CPUs are not in gaming segment. Intel promotes LGA1151 stuff for that.

Why would we compare Ryzen to Intel HEDT stuff? Because they have similar number of some weird elements called "cores"? Why not the physical size of a die?

, no DX11 vs 12

At this point everything should be tested on DX12 anyway. Why bother with both?

Ryzen using slower RAM than 7700k (after AMD said infinity fabric uses RAM clock)

Ryzen platform has compatibility issues with faster DDR at this point, Intel doesn't. We're comparing platforms' performance, not just raw design superiority.

Think about it. If it turned out that Intel chips work with every GPU available, but Ryzen (for whatever reason) is not compatible with anything faster than a GTX1050, would this be an important factor while evaluating their gaming potential?
Or would you also expect reviewers to use GTX1060 for both and totally omit the tiny "detail"?

 

[Rowsol]

4k is a pointless test for CPU testing and mostly a waste of GPU power.

AMD categorized Ryzen as gaming segment

They compared it to the 1k chip and said same performance, half price. They also tested it against a core i5 or i7 (don't remember which) while streaming, again, showing the gains of extra cores. Anyone with a brain knows a 16t chip isn't built for gaming.

 

[refillable]

You assume people actually have 4K LCDs?

What's the difference between that and assuming the otherwise for a (let's assume) GTX 1080 Ti and this CPU? Anecdotal evidence.

AMD categorized Ryzen as gaming segment. TPU is primarly a gaming-oriented site and so is the review.
In the blue camp HEDT CPUs are not in gaming segment. Intel promotes LGA1151 stuff for that.

Why would we compare Ryzen to Intel HEDT stuff? Because they have similar number of some weird elements called "cores"? Why not the physical size of a die?

Wrong. AMD categorized Ryzen as content creator processor as well. This part of your post has nothing right in it.

Think about it. If it turned out that Intel chips work with every GPU available, but Ryzen (for whatever reason) is not compatible with anything faster than a GTX1050, would this be an important factor while evaluating their gaming potential?
Or would you also expect reviewers to use GTX1060 for both and totally omit the tiny "detail"?

It isn't. What now, cherry pick Hitman, Total War and Fallout again perhaps?

And I SERIOUSLY consider giving a free i3 or Celeron for an "average customer" like you.

 

[happy medium]

AMD Ryzen 7 1800X review: what's the real story with gaming?
http://www.eurogamer.net/articles/digitalfoundry-2017-amd-ryzen-7-1800x-review

This is how you do a CPU review, you take away the gpu bottleneck as much as possible.
quote:
"We believe that a CPU purchase should last for years, so rather than test processors with a particular GPU at standard gaming conditions, we opt instead to take the graphics hardware out of the test results as best we can and to attempt to concentrate more closely on a processor's gaming potential. The aim here is to ascertain relative performance between CPUs when running game engine code - this gives a better idea of how 'lastable' a potential processor may be."

Looks like a i5 7600k trades blows with a 1800x.

Guys, Pleaase read the whole review!
http://www.eurogamer.net/articles/digitalfoundry-2017-amd-ryzen-7-1800x-review

The fact is if your NOT GPU LIMITED a Ryzen cpu is about 15/20% slower than a Intel cpu when gaming.
It even stated that a gtx1080ti overclocked is sometimes limited @ 1080p
AND if you overclock the Intel cpu's will beat a Ryzen cpu even more.

The good news is, if you are encoding movies all day , Ryzen is a good buy.

 

[refillable]

"We believe that a CPU purchase should last for years, so rather than test processors with a particular GPU at standard gaming conditions, we opt instead to take the graphics hardware out of the test results as best we can and to attempt to concentrate more closely on a processor's gaming potential. The aim here is to ascertain relative performance between CPUs when running game engine code - this gives a better idea of how 'lastable' a potential processor may be."

Has been refuted countless of times. Back in the days of Bulldozer, people were showing Civ 5 benches in 800x600 and put Bulldozers to be half as fast as their Sandy Bridge counterparts. And they simply inferred the same thing - Sandy Bridge is more "lastable", but it turns out that it's a flawed logic. Even until now those crappy faildozers (or piledrivers) can still keep up (in games) with having around 80-90% the performance of the 2500K (according to computerbase.de).

 

[notb]

They compared it to the 1k chip and said same performance, half price. They also tested it against a core i5 or i7 (don't remember which) while streaming, again, showing the gains of extra cores. Anyone with a brain knows a 16t chip isn't built for gaming.

But now suddenly we're talking about what clients should know?

AMD is selling Ryzen as general consumer / gaming solution. End of story. This is all over both the platform specs and the marketing surrounding it.
Naples is the upcoming server solution. We don't know about a workstation solution at the moment.
Go to their website. Ryzen is in desktop solutions. No CPU is mentioned in the workstation part - just GPUs:
http://www.amd.com/en/products/workstations

AMD knows very well that they can make a workstation-grade platform and ask $1000 for a CPU - just like Intel does.
Have you seen the latest rumors?
http://wccftech.com/amd-working-16-core-ryzen-cpu

Don't get me wrong: Ryzen 7's multi-thread performance is up there with Intel HEDT, but the whole platform isn't - it lacks robustness and features (ECC support being the most obvious).
Comparing Ryzen and Broadwell-E simply by looking on the performance and price is naive at least (I don't want to call it "biased" or "fanboyish"). For the extra $500 Intel sells you a lot more than the 8C/16T setup.

If you're just after performance, check the i7-6800K. It's has basically the same performance as the Ryzen 1700 and it costs "just" $400.
Sure, Ryzen 1700 is still cheaper, but that's $330 vs $400 - something we've been used to with AMD vs Intel.
And of course it's just 6C/12T going neck-and-neck with 8C/16T from AMD.

 

[Capitan Harlock]

You assume people actually have 4K LCDs?

I have a 43" 4k Lcd IPS monitor with a LG pannel XD.

 

[nemesis.ie]

Update: running tests at 3850/1.3v (Ryzen Master), CPU-Z says 1.25v. RAM @ 3200MHz, AIDA64 reporting a little over 50,000MB/sec RAM read speed.

Cracked 20,000 in CPU-Z bench.

Running with a ghetto TT copper cooler with 120mm fan on top (via a funnel) pulling air off the chip ... kind of works but the Kraken will be better whenever I get the mount.

AIDA64 reports 55c. Ryzan Master about 75 ... which ties in to the 20c offset nonsense.

Regarding ECC, I don't have any to test but all the ECC options (and indeed server platform options like socket/die RAM interleaving) are present in the UEFI settings ...

 

[refillable]

Don't get me wrong: Ryzen 7's multi-thread performance is up there with Intel HEDT, but the whole platform isn't - it lacks robustness and features (ECC support being the most obvious).

Rubbish, and if it really if some is true, relating those to gamers and content creators' perspective (which AMD clearly marketed Ryzen 7 to) is still dumb.

Comparing Ryzen and Broadwell-E simply by looking on the performance and price is naive at least (I don't want to call it "biased" or "fanboyish"). For the extra $500 Intel sells you a lot more than the 8C/16T setup.

If you're just after performance, check the i7-6800K. It's has basically the same performance as the Ryzen 1700 and it costs "just" $400.
Sure, Ryzen 1700 is still cheaper, but that's $330 vs $400 - something we've been used to with AMD vs Intel.
And of course it's just 6C/12T going neck-and-neck with 8C/16T from AMD.

Not at all. 6 cores are still too weak for me. The R7 still wins on almost all multithreaded benches out there.

 

[nem..]

AMD Ryzen 7 1800X review: what's the real story with gaming?
http://www.eurogamer.net/articles/digitalfoundry-2017-amd-ryzen-7-1800x-review

This is how you do a CPU review, you take away the gpu bottleneck as much as possible.
quote:
"We believe that a CPU purchase should last for years, so rather than test processors with a particular GPU at standard gaming conditions, we opt instead to take the graphics hardware out of the test results as best we can and to attempt to concentrate more closely on a processor's gaming potential. The aim here is to ascertain relative performance between CPUs when running game engine code - this gives a better idea of how 'lastable' a potential processor may be."

Looks like a i5 7600k trades blows with a 1800x.


Guys, Pleaase read the whole review!
http://www.eurogamer.net/articles/digitalfoundry-2017-amd-ryzen-7-1800x-review

The fact is if your NOT GPU LIMITED a Ryzen cpu is about 15/20% slower than a Intel cpu when gaming.
It even stated that a gtx1080ti overclocked is sometimes limited @ 1080p
AND if you overclock the Intel cpu's will beat a Ryzen cpu even more.

The good news is, if you are encoding movies all day , Ryzen is a good buy.

<br class="Apple-interchange-newline"><div id="inner-editor"></div>

Man since we are credulous to believe in Digital foundry (that is known for years who they serve ), why not also to believe to TOM$HARDWARE where a 6900k with ipc Broadwell can defeat the 7700k kabylake, as has not been seen In other reviews this in gamming and in 1080p, something similar to the one of DIGITAL FOUNDRY. What fools the intel engineers that even their cpu broadwell defeat to their fastest cpu kabylake (this can not be true), i guess only in titles where are used more than four cores but not at all tittles as they show it. :/


btw according to this japanese review this is the ipc of this cpus.
link http://ascii.jp/elem/000/001/445/1445029/

 

[lexluthermiester]

This is one of the worst review i've ever seen in this website: Bad tests, bad choice and very limited amount of CPU models, worse RAM speed and latency on one side(Ryzen, the most dependent)...

I truly expect an update of this review in the near future

Intel fanboy nonsense. I have read/watched MANY reviews and this one was well thought out. The tests chosen show a very clear methodology of careful, thoughtful planning with a focus on being fair to both platforms. The results and insights offered clearly show the reviewers excitement for AMD's new line of CPU's which are good performers by comparison to Intel's BEST line up and are even more competitive in pricing.

Your rather sad little opinion comes off childish at best. Go boil your head.

Oh, and for the record, I own one of the best Intel's CPUs available and will not be going to AMD. Still, no one but arrogant, prideful Intel fanboy's are going to say that AMD has not done very well with these CPU offerings.

AMD's Ryzen 5 and 3 lineup are going to be just as interesting and exciting!

 

[nem..]

RYZEN + 3600MHZ RAM - closing the 7700K gap in gaming?

 

[EarthDog]

I guess it's this video's turn to be posted 50 times at TPU, lol!

 

[medi01]

I guess it's this video's turn to be posted 50 times at TPU, lol!

I don't find it funny.
And I think we need screenshots from it, difference is MASSIVE:

 

[medi01]

Practical examples:

http://www.neogaf.com/forum/showthread.php?t=1348347&page=44

 

[EarthDog]

Its funny when you see it 10 times in 4 threads.

I'd like to see TW3 empirically tested instead of some random dude in the forum and his results. I'd also like to see that same comparison with a modern intel cpu. If the fps rates are similar, well, you can finish the story.

 

[happy medium]

I don't find it funny.
And I think we need screenshots from it, difference is MASSIVE:


View attachment 85614
View attachment 85615
View attachment 85616

View attachment 85617

View attachment 85618

View attachment 85619

Dude he is using a gtx1070, gpu bottleneck, end of story.
He needs to use a gtx1080ti overclocked to relieve the gpu bottleneck.

He also should use ddr4 4000 ram with the Intel system, it also makes a difference.

 

[medi01]

gpu bottleneck, end of story.

Oh, FFS, how is it "just a GPU bottleneck" if performance varies so wildly between CPUs?

Not to mention that 1070 outsells Titan like what, 50 to 1 and is more relevant?

 

[nemesis.ie]

I'd love to see this same 3600MHz test with 2x RX480s which would also be cheaper than 1 x 1080ti.

 

[GoldenX]

The reactions from the fanboys are truly something. Intel boys are BUT MUH GAMES FPS, while AMD boys are angry because this review or any other doesn't show Ryzen winning on everything.
Was it like this on the Athlon 64 vs Pentium 4 days?

 

[nemesis.ie]

You have to ask?

 

[notb]

Was it like this on the Athlon 64 vs Pentium 4 days?

Not really. Back then Intel and AMD made similar CPUs, with similar number of cores (1-2...). Actually back then Intel was (slightly) leading in multi-core tasks (thanks to HT).
Thing is though... the most recent CPU was the fastest. It changed all the time.

On one hand, the fanboy war was very constant, but on the other, AFAIR it was less intensive than today.
AMD guys were living in frustration for 5 years - now they finally have a Ryzon to live!
As for Intel crowd... I guess they simply missed arguing.

 

[happy medium]

Oh, FFS, how is it "just a GPU bottleneck" if performance varies so wildly between CPUs?

Not to mention that 1070 outsells Titan like what, 50 to 1 and is more relevant?

How does it matter how many people buy the faster gpu's?
Next year when a $350 mainstream upper midrange gpu is as fast as a gtx1080ti and the Ryzen cpu's are falling even more behind, I won't say I told you so.

The way to test a cpu's performance is to have no gpu bottleneck, its always been that way.
What now we make special rules for Ryzen cpu's?

 

[medi01]

How does it matter how many people buy the faster gpu's?

Because, as seen with the 1070, you actually can NOT freaking deduct how certain CPU+ GPU combination will actually work out in games?