Official AMD Ryzen Benchmarks, Reviews, Prices, and Discussion

[Mockingbird]

It's not free if you have an OEM win 7 license and get a new board.

Well, you would be buying a new copy of Windows anyway in that case.

 

[RichUK]

It's not free if you have an OEM win 7 license and get a new board.

There's a registry hack that allows you to circumvent the OEM restriction.

I used a Windows 7 license from a HP laptop from many many years ago and upgraded all the way to Windows 10 on my diy PC.

 

[utmode]

['It is worth noting that Intel, as you would expect any competitor, went through each open source project AMD had used an implemented updates to assist dead/bad code (e.g. coalescing read/writes to comply with AMD/Intel design guidelines) or offer improvements (adding in 256-bit vector codes to combine consecutive 128-bit compute*), which is why some of the results from the launch today are different from what AMD has shown. (If anyone thinks this is ‘unfair’, it begs a bigger question as to how IPC is a good measure of performance if the code is IPC limiting itself, which is a topic for another day.) At the Tech Day, AMD went into further detail regarding various internal benchmarks, both CPU and gaming related, showing either near parity or a fair distribution of win/loss metrics, but ultimately giving better price/performance.']
Any more detail about above point. BTW it is taken from Anandtech 1800x review.

 

[tamz_msc]

So the AMD CPU is much better, but still, in some respects, the same - great in multithreaded situations, barely adequate in single-threaded siluations, and bolow average (expectations) in gaming. I am glad I didn't hold my breath from my post several pages back, lol.

I hope optimizations can improve on what they have, but comments like this ""The overwhelming majority [of reviews] are highly positive on Ryzen and its performance," from John Taylor, which is very similar to what AMD was posting yesterday, don't keep me incredibly optimistic.

How is Broadwell IPC barely adequate for single-threaded performance?

 

[french toast]

['t is worth noting that Intel, as you would expect any competitor, went through each open source project AMD had used an implemented updates to assist dead/bad code (e.g. coalescing read/writes to comply with AMD/Intel design guidelines) or offer improvements (adding in 256-bit vector codes to combine consecutive 128-bit compute*), which is why some of the results from the launch today are different from what AMD has shown. (If anyone thinks this is ‘unfair’, it begs a bigger question as to how IPC is a good measure of performance if the code is IPC limiting itself, which is a topic for another day.) At the Tech Day, AMD went into further detail regarding various internal benchmarks, both CPU and gaming related, showing either near parity or a fair distribution of win/loss metrics, but ultimately giving better price/performance.']
Any more detail about above point. BTW it is taken from Anandtech 1800x review.

Yea i forgot to post about this, not sure what i think really, amd has option to do the same in games.

 

[tamz_msc]

https://www.bit-tech.net/news/hardware/2016/01/19/amd-accuses-sysbench-bias/1

Woohoo, it hit the ball out of the park in a benchmark AMD accuses of being Intel-biased!

 

[french toast]

Woohoo, it hit the ball out of the park in a benchmark AMD accuses of being Intel-biased!

add that to cinebench.

 

[piesquared]

Woohoo, it hit the ball out of the park in a benchmark AMD accuses of being Intel-biased!

And they were. That's been proven, and AMD aren't the only ones that accused Sysmark of bias. VIA and Nvidia also dropped support for BABPCO. Also SYSbench, SYSMark.

 

[unseenmorbidity]

Woohoo, it hit the ball out of the park in a benchmark AMD accuses of being Intel-biased!

Right, because that makes sense... Clearly that absolutely confirms they weren't biased in the past.

Ohh look, Jack the Ripper didn't murder someone today. That obviously absolves him of any charges of past murders!

 

[BrandonT]

I notice that Anand has the 1800X swapping places with Intel 7700K on the Handbrake tests, while other sites show a considerable advantage to Ryzen on Handbrake. This is the sole reason I'm looking to upgrade from my 3570k @ 4.4ghz. Thoughts?

Anyone have some input?

 

[R0H1T]

Anyone have some input?

Should be much better at handbrake ~

 

[zinfamous]

It will recover. Stop looking at these types of events.
The consumer investors over react and the intelligent ones exploit them.

This is the same thing with people freaking out over the Nvidia stock drop. Investors do some idiotic things like selling out of BP after the oil spill.

All this shows is that you have an opportunity to exploit some idiotic investors.
In general, investors over react, then it corrects.

So ya, exploit some people if you want. The price will recover and already has been since I started watching this morning looking for a buying opportunity once I saw the gaming benches and I know how dumb investors are.

It really is funny how dumb people can be. Zen is not really a gaming play, but what we actually see here is that AMD is actually competitive again in gaming. They will actually gain marketshare here where they haven't even stepped up to bat over the last 5-7 years. People that think Zen lives and dies on gaming have no idea what is going on. The real value for AMD with Zen has yet to be seen, but yesterday's official numbers are really as good as anyone could hope when the real market comes into play in 2nd-3rd Qtr.

The fact that they are even at the plate with gaming again is just a bonus. Exploit the fools!

 

[USER8000]

Loon told me that once it is fixed it will only be about a 2-5% difference, which is the same as intel.

Hopefully it is fixed, or at least mostly so, by the time the R3s come out.

Yeah but posts by AMD PR like this are not helping:

AMD PR guy says its not a Windows issue but a game issue,which would be far worse,since there are at least a dozen games at launch which have issues,and loads of games which might be affected which have not been tested.

However,if you look at discussion on the matter,people think it is more likely to be a windows scheduler issue anyway due to the dual CCX design of Ryzen(which possibly means 4C ones won't be affected).

Way to go AMD PR,make a problem sound worse than it might be!!

 

[tamz_msc]

Right, because that makes sense... Clearly that absolutely confirms they weren't biased in the past.

Ohh look, Jack the Ripper didn't murder someone today. That obviously absolves him of any charges of past murders!

Hey I never claimed anything about the truth of the accusations of AMD, did I?

 

[Puffnstuff]

post removed

Not only that but some of them are quick to attack any poster who feels the need to point out irregularities in the Ryzen's performance against industry standard benchmarks and measurements. I like performance period and I don't care whether the blue or the green team delivers it as I've supported both. I used to swap GPU's the same way until Nvidia established their dominance just as Intel has with CPU's. I really wanted AMD to pull a rabbit out of their hat with this product but after seeing a variety of test results I feel let down. Their stock price drop following the release demonstrates that their stakeholders feel the same way about it. I can only hope that they're able to improve the design and boost performance. As of today there's nothing about the 1800x that would compel me to want to abandon my 4790k for it.

 

[Agent-47]

I'm truly disappointed that they chose to demonstrate deceiving gaming benchmarks to promote the thing when it can barely hold its own against years old Intel offerings that are much cheaper.

there are several things to be disappointed with AMD's launch.
1. they did not allow adequate time to the MB manufacturer to iron out the BIOS and RAM support.
2. their decision to place 1700-nonX against a highly clocked 7700k for only Multithreaded performance, discarding single thread. In an ideal world, where total throughput matter, it would have been perfect, but 7700k's strength is its clock speed and the match up played into intels favor from day one
3. should have launched at least on 6c12t, as that is the true 7700k counter part.

but to claim the Zen family as a whole is a source of disappointment is truely a symptom of cognitive dissonance.
1. Intel had Ram issues in the near past.
2. its perf/W is beyond anything we saw in the past and so once the process matures, it will clock a lot better. THAT is the true indicator of how far things have moved forward with this launch. Intel has done nothing in that department for years.
3. I honestly don't think gaming is an issue. it will require a patch from the game developers to let the games identity which of the threads in a zen is logical and which is not. why are we even surprised by this is beyond me. and at the end of the day, gamers are not the users of high-end CPUs. 7700k is not a high-end CPU today. a better 7700k counterpart would be a 1600x which would be 33% cheaper.
4. IPC may not be top notch in some cases compared to intel, but it is more than made up for by perf/watt, and once the process matures, but clocks. even then the multithreaded throughout is phenomenal. and ST IPC will only improve as softwares learn to take advantage of the new architecture

 

[tamz_msc]

Yeah but posts by AMD PR like this are not helping:

AMD PR guy says its not a Windows issue but a game issue,which would be far worse,since there are at least a dozen games at launch which have issues,and loads of games which might be affected which have not been tested.

However,if you look at discussion on the matter,people think it is more likely to be a windows scheduler issue anyway due to the dual CCX design of Ryzen(which possibly means 4C ones won't be affected).

Way to go AMD PR,make a problem sound worse than it might be!!

At this point I would not exclude the possibility of the issues being both scheduling-related and due to unoptimized game code. The only thing one would wish now is that it gets resolved via a Windows update that fixes the scheduling, that would arrive in a timeframe that is less uncertain than changes to game code.

 

[Despoiler]

PostDeleted.

Or if Intel had made this chip would it be Intel fans singing praises to the high heavens for a chip that is a monster in productivity and good enough in gaming? All that while having excellent price/perf for chip that is living in two market segments in addition to having better perf/watt. See how that type of argument works? It can go in both directions which makes it a terrible argument. You might try upping your post quality given you are new here.

 

[tamz_msc]

Post Deleted.

Intel already has this CPU, alas it doesn't have a 65W TDP and unfortunately costs 1000$.

 

[piesquared]

Not only that but some of them are quick to attack any poster who feels the need to point out irregularities in the Ryzen's performance against industry standard benchmarks and measurements. I like performance period and I don't care whether the blue or the green team delivers it as I've supported both. I used to swap GPU's the same way until Nvidia established their dominance just as Intel has with CPU's. I really wanted AMD to pull a rabbit out of their hat with this product but after seeing a variety of test results I feel let down. Their stock price drop following the release demonstrates that their stakeholders feel the same way about it. I can only hope that they're able to improve the design and boost performance. As of today there's nothing about the 1800x that would compel me to want to abandon my 4790k for it.

What of the downgrades to intel's stock citing lower revenue ahead? Has there been any downgrades to AMD's?

 

[unseenmorbidity]

Not only that but some of them are quick to attack any poster who feels the need to point out irregularities in the Ryzen's performance against industry standard benchmarks and measurements. I like performance period and I don't care whether the blue or the green team delivers it as I've supported both. I used to swap GPU's the same way until Nvidia established their dominance just as Intel has with CPU's. I really wanted AMD to pull a rabbit out of their hat with this product but after seeing a variety of test results I feel let down. Their stock price drop following the release demonstrates that their stakeholders feel the same way about it. I can only hope that they're able to improve the design and boost performance. As of today there's nothing about the 1800x that would compel me to want to abandon my 4790k for it.

Comments like this do little to prove you are the objective voice of reason.

 

[Conroe]

How is Broadwell IPC barely adequate for single-threaded performance?

It only matches broadewell in MT. A well clocked 2600K matches it games.

 

[french toast]

I tried to find reviews of ryzen using the gigabyte mobos, unfortunately those reviews tend to be mainstream and so are the worst reviews ive ever read, with the reviwer unable to understand their own results! (looking at you gamespot/pc world) extremetech didnt seem to have many gaming comparisons either.
Think we are going to have to sit back and wait.

 

[AMDisTheBEST]

It will recover. Stop looking at these types of events.
The consumer investors over react and the intelligent ones exploit them.

This is the same thing with people freaking out over the Nvidia stock drop. Investors do some idiotic things like selling out of BP after the oil spill.

All this shows is that you have an opportunity to exploit some idiotic investors.
In general, investors over react, then it corrects.

So ya, exploit some people if you want. The price will recover and already has been since I started watching this morning looking for a buying opportunity once I saw the gaming benches and I know how dumb investors are.

Exploiting others sound fun....

 

[R0H1T]

Sure are a lot of AMD apologists in the forum. I have the feeling they wouldn't be so sympathetic to Intel if they had made this CPU.

Oh I'm sure there are lots of (tech) sites & corporations doing their bidding ~ Cisco shrinks: Revenue, profit and margin all dipped in Q2 2017

Kramer and Robbins were both tight-lipped about the details of the product failures allegedly caused by Intel's Atom chip-clock slip, saying only that the company's working “proactively” with customers, and doesn't expect it to cause a significant financial fallout.

Intel Atom chips have been dying for at least 18 months – only now is truth coming to light