Ryzen 2700X OCed to 4.3Ghz (1.4v) across all cores, performance numbers included.

[cj09beira]

2 minutes ago, Cookybiscuit said:

If multi-threading in games was that good Ryzen would easily outperform Intel, but it doesn't yet. There's still a ways to go before moar cores>fast cores.

PissMark isn't an indicator of anything, it's a useless test.

same can be said for your chart made with really outdated prices 

[Cookybiscuit]

Just now, Bananasplit_00 said:

oh wow missed this cheeky edit, nice way to get out of screwing up... anyway i was just saying that i didnt remember Ryzen being 30% behind, used your provided benchmark and calculated out you are claiming almost doubble what the actiual numbers are  To answer your question here, because its cheaper. the 1600 + mobo is cheaper than the I5 7400 + mobo was or about the same, so it was generally recomended and the 1700 is great value if you do something that can use the cores and game ontop of that, otherwise the Intel options can be better. 

7400 is irrelevant to the discussion when the 8400 exists and trounces the 1600.

[Dylanc1500]

25 minutes ago, jde3 said:

When did I say better?

 

It's got a novel design, it's good for stuff.. its darn good for my use case. But better? What is that metric? Best ever? no.. what are we doing? x86 isn't even the best architecture.

Someone should develop doom for Power, I'll run it on Power 9 for giggles. Then we will have a comparison of architectures (as far as ONE individual game is concerned).

[cj09beira]

1 minute ago, GoldenLag said:

They are okay for minor arcitectural changes. They still havent fixed the clock issue which seems to be linked to the arcitecture and not so much the node. We got exactly what was expected from the 14-12 nm shrink.

pretty sure its a node issue, this is still a refinement to a low power node

[jde3]

3 minutes ago, Dylanc1500 said:

Someone should develop doom for Power, I'll run it on Power 9 for giggles. Then we will have a comparison of architectures (as far as ONE individual game is concerned).

The original doom ran on Irix and SGI systems.

[Monarch]

31 minutes ago, hobobobo said:

Im upset since its unrepresentative. 720 benachmark highlights latency difference mainly, where ryzen is behind 20-30%. Show me someone gaming on 720p monitor with a 8700k and a vega64 and ill know how an oblivious idiot looks like

 

Benchmarking at 720p is done to ensure a CPU-bound scenario at all times throughout the testing, regardless of what's going on a game. Wasting hours to find a real CPU-intensive part of the game makes no sense, and even if you do find it you can't rerun the same test with different CPUs in most games due to a lot of variables.

[Dylanc1500]

1 minute ago, jde3 said:

The original doom ran on Irix and SGI systems.

That is very true. However I don't think anyone here really cares about how well the original doom runs nowadays. Then again there could be some outliers that might care.

[GoldenLag]

7 minutes ago, Cookybiscuit said:

PissMark isn't an indicator of anything, it's a useless test.

Well, the 720p benchmark got thrown out of the window then. If synthetics cant be used for comparing CPU's. Why should the 720p benchmark count? They are both supposed to utilize the CPU to its maximum. 

[Coaxialgamer]

1 minute ago, Dylanc1500 said:

That is very true. However I don't think anyone here really cares about how well the original doom runs nowadays. Then again there could be some outliers that might care.

#DOOMONARDUINO

[Master Disaster]

5 minutes ago, Monarch said:

 

Benchmarking at 720p is done to ensure a CPU-bound scenario at all times throughout the testing, regardless of what's going on a game. Wasting hours to find a real CPU-intensive part of the game makes no sense, and even if you do find it you can't rerun the same test with different CPUs in most games due to a lot of variables.

We understand why it's done, the point is it's not representative of real world performance and therefore meaningless.

 

Just like when VW say their latest diesel can do 90 mpg, sure in a lab on a rolling road with no variables at all but get it out on the road and your doing good to hit 60. Meaningless.

[leadeater]

4 minutes ago, Cookybiscuit said:

If multi-threading in games was that good Ryzen would easily outperform Intel, but it doesn't yet. There's still a ways to go before moar cores>fast cores.

And if the difference actually had anything to do with the basic core performance and not 10+ years of optimization specifically for Intel architectures then yes that would be true. If you look at games that actually claim to be well optimized for multiple cores then there is next to no difference.

 

As for my GalCiv 3 example absolutely yes a Ryzen 1600 or better would crush every 4 core Intel CPU in that game. If you play a lot of RTS and TBS games like I do you'll know once you hit late game turn times and general game interaction gets SUPER bad and only the most well optimized multi core friendly games pass muster here, rest have to go right in to the "click next then walk away and do something else" play style and is why Civ6 disappointed me so much.

 

It's like everyone expects a chef to get a Michelin star on the first try, a dish doesn't need one to still be a great meal.

[jde3]

Intel is a really spooky company btw. You all of course remember Palladium in the Itanium chips.. thank god we dodged that one.. prob due to the Athlon64 being so good. This was real kiddos. https://www.theregister.co.uk/2002/06/25/why_intel_loves_palladium/

 

Win-tel wanted to make it so the only thing that could run on your CPU had to be approved software by them.

 

This is the shit we were dealing with when Microsoft and Intel had no opposition.

[GoldenLag]

Thank god AMD is here. Even if Intel got to 6 core. Its gonna be a decade of stagnation and 10nm+++++ before 8 core would have become mainstream

[Dylanc1500]

1 minute ago, Coaxialgamer said:

#DOOMONARDUINO

Better be overclocking.

#ARDUINOLIVESMATTER.

[Princess Luna]

3 hours ago, leadeater said:

Call me when I care about CS:GO and when monitors can do over 240Hz 

As someone who plays Counter-Strike since 1.5 I vouch for this video

[laminutederire]

3 hours ago, leadeater said:

Call me when I care about CS:GO and when monitors can do over 240Hz 

 

I won't be expecting a call though since I don't care about CS:GO and never will lol.

From someone who cares about this game, it saturates around 144-165Hz anyway. 240Hz is mostly placebo with this game anyway.

(And it's just so badly optimized sometimes that it shouldn't be used in any arguments )

[Monarch]

Just now, Master Disaster said:

We understand why it's done, the point is it's not representative of real world performance and therefore meaningless.

How is it not representative of real world performance? Instead of finding a CPU intensive part of the game and test CPUs there, you create a CPU intensive situation yourself. Of course, you are increasing the amount of fps by doing this which isn't representative of the actual fps you'd get in an actual cpu heavy part of the game, but the the performance difference between the cpus is what matters.

 

 

 

[Master Disaster]

2 minutes ago, Monarch said:

How is it not representative of real world performance? Instead of finding a CPU intensive part of the game and test CPUs there, you create a CPU intensive situation yourself. Of course, you are increasing the amount of fps by doing this which isn't representative of the actual fps you'd get in an actual cpu heavy part of the game, but the the performance difference between the cpus is what matters.

 

 

 

Because out in the real world no one actually plays at 720p.

 

The resulting data presents the best possible outcome and not necessarily the real world outcome.

[GoldenLag]

22 minutes ago, cj09beira said:

pretty sure its a node issue, this is still a refinement to a low power node

Would be nice to test cpu's on different nodes. How would Intel fare on AMD's nodes and how AMD would fare on theirs. 

[Curufinwe_wins]

2 minutes ago, Master Disaster said:

Because out in the real world no one actually plays at 720p.

 

The resulting data presents the best possible outcome and not necessarily the real world outcome.

Ok, so I dont have a horse in the game here, but let's just take a step back. So what it does is test what happens in CPU bound gaming loads, which is not the same sort of thing as CPU bound cinebench or etc.

 

Thing is. Right now it seems ludicrous. No one plays at 720p. Ok. So what happens when you take a 1080ti or a 1280ti (years down the road) and play at 1080p. More or less the exact same thing. Highly CPU bound. And yeah, I know idiots that bought 1080tis to run at 1080p. They do exist. But even then, suppose you really dont think anyone will run at less than 1440p... at the current rate of improvement, how long before current games at 1440p will be cpu bottlenecked by GPU advancements? 

 

Hell, I can show you CPU bottlenecking on gta5 at 4k from a 4790k at 4.7 GHz moving to a 5930k at the same speed just from two 980tis...

[leadeater]

5 minutes ago, Princess Cadence said:

As someone who plays Counter-Strike since 1.5 I vouch for this video

2 Jokes:

Sorry I work for Ubisoft

Counter-strike..... IGNORED

 

 

On topic, yea it's not like I disagree with that however a lot of that smoothness comes from frame lows being higher rather than specifically the average being higher. As a gamer who thinks games that look anything like minecraft should burn in the deepest fires of hell anything above 120 FPS is wasted potential.

 

When I was playing cod4 my target was 100, which meant turning a lot of stuff down and going lower res on an X800. Still my most fun PC gaming years.

[Cookybiscuit]

22 minutes ago, leadeater said:

And if the difference actually had anything to do with the basic core performance and not 10+ years of optimization specifically for Intel architectures then yes that would be true. If you look at games that actually claim to be well optimized for multiple cores then there is next to no difference.

 

As for my GalCiv 3 example absolutely yes a Ryzen 1600 or better would crush every 4 core Intel CPU in that game. If you play a lot of RTS and TBS games like I do you'll know once you hit late game turn times and general game interaction gets SUPER bad and only the most well optimized multi core friendly games pass muster here, rest have to go right in to the "click next then walk away and do something else" play style and is why Civ6 disappointed me so much.

 

It's like everyone expects a chef to get a Michelin star on the first try, a dish doesn't need one to still be a great meal.

Intel CPUs don't have four cores anymore though

[XenosTech]

2 hours ago, Cookybiscuit said:

It does the job, you're talking about a seven year old product still being relevant, in the tech industry that's an extremely long lifespan.

If you said that about a GPU sure but CPU's not so much.... CPU's tend to out last GPU's in usefulness by a long shot

[Monarch]

25 minutes ago, Master Disaster said:

Because out in the real world no one actually plays at 720p.

 

But I've just literally explained to you, in detail, why that's irrelevant. You may not play at 720p, but you will run into a CPU intensive part of the game in which case your framerate depends on the CPU, just like when you create a CPU-bound situation to represent a real one by lowering res to 720p. Say a difference between an i5 and i7 in a real world CPU-bound situation is 30%, you should have about the same performance difference in a 720p CPU-bound scenario you created yourself to not have to look for a real one in the game.

[hobobobo]

7 minutes ago, Cookybiscuit said:

Intel CPUs don't have four cores anymore though

how many cores is 8300 or 8350k?)