Again IPC has barley nothing to do with a CPU's performance. Single core performance and clock speed don't scale and the industry has moved away from that modle. All you're proving is if you use that set of tests with that hardware configuration you can produce that bar chart. That's great.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
5Ghz Amd & Intel (speculation topic)
- Thread starter geordiejc
- Start date
Again IPC has barley nothing to do with a CPU's performance. Single core performance and clock speed don't scale and the industry has moved away from that modle. All you're proving is if you use that set of tests with that hardware configuration you can produce that bar chart. That's great.
Ram was capable of more than 2400mhz on day one, plus as we've seen the gains from 2666-3200 aren't as big as AMD would have you believe.
Do you think that intel wouldn't get a boost from faster ram also?
Ryzen is not on par with sky/kaby/coffee with IPC no matter how much you try to dress it up.
We have so many AMD apologists on this forum, whenever a benchmark get shown we hear the following;
Ram not at 3200
Early bios
Not optimised for AMD
Reviewer is a shill
Test is irrelevant
Try both and see how it is.
IPC has nothing to do with CPU performance? If both CPU's are clocked the same then IPC has EVERYTHING to do with performance.
A 5ghz fx would get mullered by a 5ghz kaby, is that not down to IPC?
I'll leave this here.
Yeah probably for the best. See you in three pages...When you argue a plus 1Ghz increase in clock speed = the same performance because of of few percent difference in IPC.
Take a look at the link I provided, both CPU's clocked at 3.5ghz and zen loses. In fact the stilt has kaby at 35% ahead of zen in IPC over the many tests he ran
How much simpler do you want it?
That's great. The points you made have nothing to do with discussion and it's totally inaccurate and screwed view you have but thank's for the bar charts. Now I see how increasing a chips clock speed by 25% yields the same performance.
Yep just ignore everything else and focus on that. Single core brute force is all that matters and IPC will save the day. Please accept my apology as your bar charts have now proved my understanding of the situation is wrong. Could you please advise me what bar charts I need to produce to fix the performance issues I see or give me some advice on how best to interpret the problem so it goes away?
I think this is a pretty fair take on Ryzen after updates/better memory etc.
Considering at the top end there you've got a 5.1 ghz 7700k vs a 1700x @ 3.9 ghz and there's often less than 10% (fps) in it (when the 7700k has a 20% ghz advantage...) It's one that can be pulled apart in loads of directions but at least there's a good spread of stuff there with specifics.
Absolute single core isn't particularly relevant unless you're into stuff that's years old. There's not much that's TRULY multi-thread aware but... there's plenty that use more than 1 effectively.
Stolen and re-hosted below for an example.
Considering at the top end there you've got a 5.1 ghz 7700k vs a 1700x @ 3.9 ghz and there's often less than 10% (fps) in it (when the 7700k has a 20% ghz advantage...) It's one that can be pulled apart in loads of directions but at least there's a good spread of stuff there with specifics.
Absolute single core isn't particularly relevant unless you're into stuff that's years old. There's not much that's TRULY multi-thread aware but... there's plenty that use more than 1 effectively.
Stolen and re-hosted below for an example.
Detailed I think is the word you're after
check the link, there's a lot more detail and a good spread of games checked.Detailed I think is the word you're after
What graphics card got used? Can't seem to find the link.
I think this is a pretty fair take on Ryzen after updates/better memory etc.
Considering at the top end there you've got a 5.1 ghz 7700k vs a 1700x @ 3.9 ghz and there's often less than 10% (fps) in it (when the 7700k has a 20% ghz advantage...) It's one that can be pulled apart in loads of directions but at least there's a good spread of stuff there with specifics.
Absolute single core isn't particularly relevant unless you're into stuff that's years old. There's not much that's TRULY multi-thread aware but... there's plenty that use more than 1 effectively.
Stolen and re-hosted below for an example.
Almost certain there's a GPU bottleneck there.
This is why decent reviewers do 720p to eliminate the gpu as the limiting factor. Volta will change the way these graphs look.
"
Hardware Used
Core Components (Unchanging)
- NZXT 1200W Hale90v2
- For DDR4 platforms: Corsair Vengeance LPX 32GB 3200MHz*
- For Ryzen DDR4: Corsair Vengeance LPX 3000MHz clocked to 2933MHz (See Page 2)
- Premiere & Blender tests do not exceed 8GB DRAM. Capacity is a non-issue for our testing, so long as it is >16GB
- For DDR3 platforms: HyperX Savage 32GB 2400MHz
- Intel 730 480GB SSD
- Open Air Test Bench
- Cooler #1 (Air): Be Quiet! Dark Rock 3
- Cooler #2 (Cheap liquid): Asetek 570LC w/ Gentle Typhoon fan
- Cooler #3 (High-end): Kraken X62"
edit:
GPU: GTX 1080 FTW1
Right ^^ so what GPU did he use? 
I'm assuming a 1080TI?
Yes
, tho i'm not sure why he insist on using .1% lows, a small read on the drive can knock those down and, well... predictably his .1% lows are all over the place, you have the same overclocked CPU's with worse .1% lows than the stock one.
I'm assuming a 1080TI?
Detailed I think is the word you're after
Yes
, tho i'm not sure why he insist on using .1% lows, a small read on the drive can knock those down and, well... predictably his .1% lows are all over the place, you have the same overclocked CPU's with worse .1% lows than the stock one.But... what relevance is 720p?
"With a rig no-one would ever consider putting this hardware towards using, THESE ARE THE SUPER IMPORTANT RESULTS"...
You can artificially create whatever arena you want if you MUST have certain stuff "win". I think top end CPU's with a decent graphics card at 1080p is already stretching it a bit (300+ fps at 1080p is serious business... somewhere... I guess).
Ya know..?
At a point you're throwing a... probably £1200 rig at it, 1440p+ seems completely reasonable, with a look at what you'd get if you ever bothered with low end (1080p) graphics.