So what's the verdict so far on the Ryzen Processors?

[subbytna]

Intel has been challenged many times during history of x86 CPUs and got into its current position only through lots of at the least very questionable and many times criminal means, which should have resulted in upper management getting thrown to jail couple times.
http://jolt.law.harvard.edu/digest/intel-and-the-x86-architecture-a-legal-perspective
In fair and honest competition Intel would have never gotten into its position of last ten years.

9900K's eight core processing power getting matched at 50W lower power consumption tells that Intel isn't going to enjoy single thread performance advantage long.
With that kind power consumption advantage AMD has good room for tweaking boost clocks.
Even if TSMC's 7nm node doesn't mature some between making that engineering sample and mass production.
And with chiplet design and room and position for another chiplet confirmed it's pretty much guaranteed that AMD will keep pushing core counts.
Which they can do easily while staying inside 9900K's true power consumption.

Sure 9900K is now the fastest desktop platform CPU.
But for its ludicrous pricing it will be insanely lot worser at holding value than all those, quaranteed to stay high end for many years, £300 Intels of the earlier decade.

First post shill?

 

[MasterOC]

I just leave my 2700 at stock gaming. According to SVI2 it only uses about 45W.

https://i.imgur.com/ghDT2My.png

I had a single core Sempron that used 45W.

Oc'ed, though, to 4.1GHz uses as much as the X version. About 135W.

Stock … oh no no no. no cpu should ever be left stock ….
but absolutely though , its a ridiculously small amount of power its sipping at stock config , but my purchase at the time of release was almost 50 quid cheaper then the x variant and mine is clocked to 4.1 @ 1.35v ( pretty much the max I can get out of it without 1.4+ v . it is let down by my motherboard choice however, never buy TUF , its now not what it used to be. at stock it bottlnecks my GPU to some degree even in highly threaded workloads.

 

[chrcoluk]

Rroff if TDP is not do with power usage, then why does intel's own XTU program show TDP based throttling when the cpu power exceeds the TDP limit?

 

[Kei]

I currently have a 2700x. I ‘upgraded’ from an i7 7700k. Truth be told it’s a downgrade in most every day scenarios. The system feels less snappy, Lightroom and photoshop are faster in batch processing with the AMD chip but it’s still a go make a cup of tea job with 8 cores as it is with four. Both Lightroom and photoshop are now slower in routine usage such as to open a file or even simple tasks like cropping. I’m not saying the AMD chip is slow but that ‘snap to’ immediacy of the i7 just isn’t there.

As for gaming we’ll im limited to 60hz so 2700x hasn’t shown me up here.

Seriously, the difference between them should be so small, that it should barely be noticeable. I moved from an i7 4820K to a threadripper 1920x both overclocked and the difference was barely noticeable in most tasks. According to puget systems testing, in most workloads there's barely 0.5 seconds between a 2700x and an 8700k. Using their simplified benchmark numbers the differences look much bigger, but looking at the actual time in seconds, it's splitting hairs.

To answer the OP's question from an owner, it does have it's cases where it just doesn't run as well as it should. GTA 5 is a good example of this where my gpu usage is pretty poor and I see fps dropping down but not to unplayable levels. I've had no problems running any games at perfectly playable framerates above 60. In my case, I knew what I was buying into and knew that it wasn't going to be at the bleeding edge with games. Pro applications run well and it flies through video conversion. Essentially, at 4GHz it's a smidge faster than my 4820k was at 4.6GHz for single threaded tasks and up to 3x faster on multi-threaded tasks. This is 1st gen, 3rd gen should see a nice big bump. Biggest plus point for me was the pcie lanes that it comes with vs the frankly appallingly nobbled intel offerings. I have 16x x2 in use 4x x1 and 3 1x at the moment and this will only increase as I add more nvme storage.

 

[Brazo]

Well, you are wrong because Windows needs as many cores as possible to run applications simultaneously. Also, the Zen cores are not weak. How did it come to you? It is the old trick by intel and adobe in which they cheat the AMD users.

That’s twice now you’ve disagreed with me and then agreed with me by your last sentence

I didn’t say weak, I said ‘weaker’

It’s kind of irrelevant what it is, this AMD/adobe ‘trick’ you speak of well I’ve never heard of it but even if your right your agreeing that such apps run sub par on AMD products and therefore supporting my own experience.

 

[4K8KW10]

I didn’t say weak, I said ‘weaker’

Zen cores are not weaker. They are the same, just more energy efficient, while intel overclocks theirs on stock high, and you pay the corresponding dissipated heat and electricity prices.

 

[Rroff]

Rroff if TDP is not do with power usage, then why does intel's own XTU program show TDP based throttling when the cpu power exceeds the TDP limit?

It is to do with power usage but not in the way people are understanding it - there is a relationship there as the amount you need to be dissipating will always be proportional to the power draw but the Intel stated TDP is not a measure of power draw directly.

 

[Martini1991]

Zen cores are not weaker. They are the same, just more energy efficient, while intel overclocks theirs on stock high, and you pay the corresponding dissipated heat and electricity prices.

Performance is weaker per core on zen due to the lower clocks.

I don't know why you're so pro AMD to the point you're lying.

 

[PiKe]

Performance is weaker per core on zen due to the lower clocks.

I don't know why you're so pro AMD to the point you're lying.

Its actually nothing to do with clocks. Its IPC which is something different.

 

[Martini1991]

Its actually nothing to do with clocks. Its IPC which is something different.

The IPC difference these days is tiny.

The clock speed difference is literally the defining factor in their performance differences per core.

 

[PiKe]

IPC clock-for-clock test.

 

[Martini1991]

I don't actually understand the point you're making.

The video you've posted doesn't specifically compare actual core by core. For example the results where the lower core count Ryzen is quite low down, it has the same IPC as the higher core count one but the graph doesn't look that way with it being an overall result.


Either AMD cores have higher IPC (They don't) or Intel cores do.

Either way there's about a 20% potential clock speed difference which is where most of the gains are per core.

 

[nashathedog]

Nope.

A very informative answer.

 

[4K8KW10]

Either AMD cores have higher IPC (They don't) or Intel cores do.

The IPC is equal which means the Zen core is as good as it can be. Now ask the manufacturing site to optimise their processes, so you get better clocks on top.

 

[ShrimpBrime]

I'm pretty sure the 9900k isn't the fastest processor hands down, as you put it. I am sure there are plenty of workloads that I use where my 1950x will hand it a good old spanking.

In fact some of what I need my cpu for wouldn't even be possible on a 9900k.

32 threads AMD vs 16 threads Intel? Umm, I don't follow. You have a low base clock and high thread count? It's worse than a 2700x for gaming and 3D benching?

Or is the 1950x superior because it does well with just Cinebench?

 

[Vince]

32 threads AMD vs 16 threads Intel? Umm, I don't follow. You have a low base clock and high thread count? It's worse than a 2700x for gaming and 3D benching?

Or is the 1950x superior because it does well with just Cinebench?

You said the 9900k was the fastest cpu period, it isn't, it's fast in gaming workloads. What's there to follow? You posted this...

If you want the FASTEST processor, you buy an Intel i9 9900K today and it's the fastest all around hands down by fact.

And I was pointing out that this is in fact total rubbish and not "fact" as you are pointing out. In my vm workloads where I require I/O and many cores the i9 9900k wouldn't be a great choice. Even if it is a great cpu.

You also show your understanding or lack of by thinking that a cpu such as the 1950x is good just for cinebench, which to be fair is a valid workload but not the only valid workload where a cpu with a higher core count would be beneficial.

 

[ShrimpBrime]

OK I see your point. I stand corrected. The 1950X is faster than the i9 9900 with multi-threaded tasks.

7980X is faster than a 1950X for multi-threaded tasks.

 

[chrcoluk]

I think I am going to disable CPB on my 2600X now running ESXi, its running power is almost halved with it disabled with probably only a 10% performance loss. (since I dont think ESXi is using XFR properly due to an intel centric scheduler). The system will be migrated to proxmox at some point so I get a more ryzen friendly scheduler. These chips in stock config (XFR disabled) are the most efficient chips I have seen for a long time.

 

[DarrenM343]

OK I see your point. I stand corrected. The 1950X is faster than the i9 9900 with multi-threaded tasks.

7980X is faster than a 1950X for multi-threaded tasks.

.....and the 2990WX is faster I think then the 7980X at those same tasks (32 cores vs 18)
Also have to consider the price of course .

 

[Vince]

.....and the 2990WX is faster I think then the 7980X at those same tasks (32 cores vs 18)
Also have to consider the price of course .

Of course, in my book it's all about what you really need it for. Would you buy 1950x/2950x or 2990wx for just a gaming machine? Probably not... but if you are some content creating streamer or some rubbish it could be a decent option. I have a game on my 1950x almost daily and paired with Radeon 7, a 4k screen and 64gb of ddr4 it can chew through work workloads while I game on the side. Win, win in my book.

For me when I really need all 16/32 cores/threads and am firing up large environments I couldn't imagine now going with anything with less grunt or with half the cores. I just wouldn't be able to do what I need the machine for on less.