Intel vs AMD Benchmarks

Forums › Desktop computing › Intel vs AMD Benchmarks

networkn

Networkn
32862 posts

Uber Geek
+1 received by user: 15453

ID Verified

Trusted

Lifetime subscriber

#280165 30-Nov-2020 15:19

https://cpu.userbenchmark.com/Compare/AMD-Ryzen-7-5800X-vs-Intel-Core-i7-10700KF/4085vsm1171560

https://cpu.userbenchmark.com/Compare/AMD-Ryzen-5-5600X-vs-Intel-Core-i5-10600K/4084vs4072

Based on these, Intel are either ahead, or slightly below but at a reasonably lower price. I got the impression from recently summary reviews I glanced over that the new Ryzen 5000 were so far ahead of Intel, why would you ever buy Intel.

I understand it's just one site, but the number of benchmarks for either chip seem to indicate it's not an anomoly.

1 | 2 | 3

20858 posts

Uber Geek
+1 received by user: 5350

Trusted

Lifetime subscriber

#2613432 30-Nov-2020 15:35

A few months ago Ryzen was ahead, but they always leapfrog each other. Doesn't really matter which you get, I think. Both are fast, and price is comparable.

One thing I've noticed with my Ryzen PC, currently back at PBTech as it never worked properly and failed within two days, is Ryzen run hotter due to AMD design philosophy. Ryzen throttle at 95 degrees and will keep getting given more power if it will make things faster until it gets there. Intel I think throttles a bit earlier. That makes Ryzen a bit more peaky and potentially noisy as the fans work harder and go up and down. There's a feature you can disable to prevent that happening, precision boot overdrive. Articles here, here.

I waited about 6 months for Ryzen / Zen 3, 5600X. I kinda wish I'd gone with Intel now, given the Ryzen PC appears to be a lemon. They'll probably fix it though.

networkn

Networkn
32862 posts

Uber Geek
+1 received by user: 15453

ID Verified

Trusted

Lifetime subscriber

#2613434 30-Nov-2020 15:37

Not sure how you mean they leapfrog? Intel hasn't released any new chips since Ryzen 5000 as far as I am aware?

timmmay

20858 posts

Uber Geek
+1 received by user: 5350

Trusted

Lifetime subscriber

#2613439 30-Nov-2020 15:45

networkn:

Not sure how you mean they leapfrog? Intel hasn't released any new chips since Ryzen 5000 as far as I am aware?

Might be that they just happen to be fairly even now, but with development cycles usually it's one leading, one trailing, and then the trailing one comes out with a new model that's a bit better.

The midrange or higher CPUs are all so fast now that the difference is largely theoretical for most people. Servers and cloud, sure they can use more performance, but a 5+ year old CPU is probably fine for most people who just surf the web.

MaxineN

Max
2049 posts

Uber Geek
+1 received by user: 1662

ID Verified

Trusted

Subscriber

#2613440 30-Nov-2020 15:46

In my personal experience. Intel CPUs throttle at 100c. Intel Ark also states that its the T-junction max so yes it will keep going until it hits it then backs off multiplier and voltage on most CPUs.

Ryzen yes runs a little warmer but with my 2600 it won't go above 65c in real world gaming and it is over clocked and over bolted.

Also I don't like user benchmark because people could have different software running in the background skewing results.

I do prefer anandtech and they have results where the 6 core 12 thread part is destroying a 10900k

https://www.anandtech.com/show/16214/amd-zen-3-ryzen-deep-dive-review-5950x-5900x-5800x-and-5700x-tested/22

Architecture comparison. AMD wins hands down and has ryzen to the top.

Ramblings from a mysterious lady who's into tech. Warning I may often create zingers.

networkn

Networkn
32862 posts

Uber Geek
+1 received by user: 15453

ID Verified

Trusted

Lifetime subscriber

#2613443 30-Nov-2020 15:49

I guess with user benchmark, one could argue that so many results, variance due to load would end up averaged out, and probably reflects "real world" performance.

To some degree, FPS based benchmarks are sort of irrelevant, so long as you can get over 120FPS, it's pretty much pointless. You can't see 200, or 400 frames per second.

Looking at the office and productivity type benchmarks performance to price ratio makes it pretty even, perhaps with Intel even slightly ahead.

ratsun81

516 posts

Ultimate Geek
+1 received by user: 183

#2613449 30-Nov-2020 15:51

Userbenchmark is not a site anyone should use to check for which CPU to buy.

The site has been outed quite a number of times of late as being incredibly biased towards intel, so much so that when Ryzen chips were released the benchmark stats they use REDUCED the weight of the results of threaded applications. Where the current theme in software development is moving towards multi-threaded performance.

In short they saw AMD doing well and beating the intel scores on their site so decided to modify the scores based off of single core vs multicore results.

Avoid

Quic Broadband

Use R212389ELFLL2 promo code for free setup at checkout.

Apple TV

Stream your favourite shows now on Apple TV (affiliate link).

MaxineN

Max
2049 posts

Uber Geek
+1 received by user: 1662

ID Verified

Trusted

Subscriber

#2613451 30-Nov-2020 15:52

Also that real world comparison is not representative of anything. Your usage will vary between someone else's. Settings will differ and so will ram speeds and timings also skewing the results from user benchmark.

Ramblings from a mysterious lady who's into tech. Warning I may often create zingers.

Kookoo

869 posts

Ultimate Geek
+1 received by user: 407

Trusted

#2613453 30-Nov-2020 15:53

I would be surprised if it was anything else to be honest.

UserBenchmark are well-known Intel fans, for better or worse.

They do an amazing job in SEO terms so that anyone looking for a quick benchmark comes across their website first. But the reliability and trustworthiness of their testing is questionable, and the tests focus on configurations and features that favour Intel over AMD. I'd say the following - if you're looking for gaming performance, you can rely on UB only as far as the specific games with the specific settings that they've tested. If you play different games, or on different settings - UB results are pretty much meaningless.

Hello, Ground!

Jase2985

13730 posts

Uber Geek
+1 received by user: 6202

ID Verified

Lifetime subscriber

#2613459 30-Nov-2020 16:03

https://www.youtube.com/watch?v=43pa3Y4Nqa8

the R5 5600x pretty much the best all round CPU at the moment

also remember the 10600k ships without a cooler so you need to factor that in when you do a cost comparison.

mentalinc

3384 posts

Uber Geek
+1 received by user: 1023

Trusted

#2613460 30-Nov-2020 16:05

Good watch on CPU to consider for workloads, note there is only 1 intel for the joke award!

https://www.youtube.com/watch?v=43pa3Y4Nqa8

or the article

https://www.gamersnexus.net/guides/3589-best-cpus-of-2020-so-far-gaming-production-overclocking-budget

CPU: AMD 5900x | RAM: GSKILL Trident Z Neo RGB F4-3600C16D-32GTZNC-32-GB | MB: Asus X570-E | GFX: EVGA FTW3 Ultra RTX 3080Ti| Monitor: LG 27GL850-B 2560x1440

Quic: https://account.quic.nz/refer/473833 R473833EQKIBX

Dial111

979 posts

Ultimate Geek
+1 received by user: 656

#2613461 30-Nov-2020 16:05

ratsun81:
Userbenchmark is not a site anyone should use to check for which CPU to buy.

The site has been outed quite a number of times of late as being incredibly biased towards intel, so much so that when Ryzen chips were released the benchmark stats they use REDUCED the weight of the results of threaded applications. Where the current theme in software development is moving towards multi-threaded performance.

In short they saw AMD doing well and beating the intel scores on their site so decided to modify the scores based off of single core vs multicore results.

Avoid

Came here to say this.. Userbenchmark is not a reliable source given their history.

Lenovo

Shop now for Lenovo laptops and other devices (affiliate link).

timmmay

20858 posts

Uber Geek
+1 received by user: 5350

Trusted

Lifetime subscriber

#2613463 30-Nov-2020 16:09

Jase2985:

the R5 5600x pretty much the best all round CPU at the moment

also remember the 10600k ships without a cooler so you need to factor that in when you do a cost comparison.

5600X cooler is fine for light use, but it makes an annoying noise. I suspect a good proportion of people who buy the 5600X will replace it.

Jase2985

13730 posts

Uber Geek
+1 received by user: 6202

ID Verified

Lifetime subscriber

#2613465 30-Nov-2020 16:14

timmmay:

5600X cooler is fine for light use, but it makes an annoying noise. I suspect a good proportion of people who buy the 5600X will replace it.

i noticed zero annoying noise on the one i setup yesterday. was quiet as a mouse and never ramped up to full speed even under a long sustained load. (Prime 95)

networkn

Networkn
32862 posts

Uber Geek
+1 received by user: 15453

ID Verified

Trusted

Lifetime subscriber

#2613466 30-Nov-2020 16:15

timmmay:

Jase2985:

the R5 5600x pretty much the best all round CPU at the moment

also remember the 10600k ships without a cooler so you need to factor that in when you do a cost comparison.

5600X cooler is fine for light use, but it makes an annoying noise. I suspect a good proportion of people who buy the 5600X will replace it.

And the 5800X doesn't have a cooler either, so that's something else to take into account.

networkn

Networkn
32862 posts

Uber Geek
+1 received by user: 15453

ID Verified

Trusted

Lifetime subscriber

#2613467 30-Nov-2020 16:17

Dial111:

Came here to say this.. Userbenchmark is not a reliable source given their history.

Be that as it may, results from other sites, for productivity and benchmarks indicate marginal differences in speed, with a fairly decent price difference, so it's not wildly off what most sites I have seen are reporting.

FPS might be one way to measure if something is hugely more powerful, but beyond 120FPS, there isn't a lot of point in more power.

1 | 2 | 3