9800x 3d vs 12900k - Battle of the Century

[JustBenching]

I don't know if it's a big ask but do you perhaps have kingdom come installed?
I'm doing research on my next cpu upgrade and a thread came up about something called "amdip" which seems to be a bit of a controversial topic.

He runs trough a specific gate that's causing him a dip that supposedly intel doesn't suffer from.
If you have time could you perhaps go in the same spot in-game with your 9800x3d and see if you get the same dip?I was told the man in the video doesn't know how to tune amd cpu's and that its's probably flck related.
Thanks in advance.

Everything dips there. Intel included. Ill upload a video of it.

Also please don't link that grifter. He is running 4800mhz ram on the 9800x 3d....

 

[kDude]

Everything dips there. Intel included. Ill upload a video of it.

Also please don't link that grifter. He is running 4800mhz ram on the 9800x 3d....

Yea sorry.I know the man has bit of a bad reputation.I only linked because I needed to timestamp the video at the part where he was showing/playing the issue at that certain spot ingame.

 

[JustBenching]

Yea sorry.I know the man has bit of a bad reputation.I only linked because I needed the timestamp the game he was playing at a specific ingame spot.

Well he doesn't know how to test anyways. KCD is heavy on the CPU because of the draw distance bars in the settings. Obviously he is running with everything low hence his framerate. What's the point of testing like that? Im currently uploading a video with those bars maxed out which actually makes it cpu heavy.

 

[phanbuey]

Bro... He literally counts how long he thinks it's taking all the assets in map to load -- out loud.

And then is like "Intel loaded it in 3 seconds... AMD loaded it in 7 seconds... it's probably due to the fact that it's still a chiplet design vs a unified".

Yes yes - the 80GB/s fabric is the reason your game took 4 seconds longer to load ~8GB of assets

 

[JustBenching]

This is the problem, he claims he is running 6400c30 "MAX OC", and yet his factorio run is doing 4800 jedec ram. Obviously he is trying to sell his craptorlakes and his 1k$ a pop consults..

Spoiler

Yea sorry.I know the man has bit of a bad reputation.I only linked because I needed to timestamp the video at the part where he was showing/playing the issue at that certain spot ingame.

Here are my runs, Low settings like he is using in his video

Spoiler

And ultra settings which is how you are supposed to test

Spoiler

 

[phanbuey]

Makes sense -- that's why his analysis has no frametime plots or any real data... it's just him running around an MP arena being like "OH LOOK THERE'S A DIP!!! IT STILL DIPS!!! DID YALL SEEE THAT... HOLY CRAP DUDE"

If he actually put a plot up you would see 95% of the time the 14900 MAXOCWTFBBQ chip is getting stomped and dipping lower. I gotta say tho -- the beef he has with Steve from HW is one the funniest things ive ever seen.

I think that last shot in the 9800X3D review where steve was like "I don't care how good you think your memory tuning skills are, you're not catching this" was referncing jufus.

 

[JustBenching]

Makes sense -- that's why his analysis has no frametime plots or any real data... it's just him running around an MP arena being like "OH LOOK THERE'S A DIP!!! IT STILL DIPS!!! DID YALL SEEE THAT... HOLY CRAP DUDE"

If he actually put a plot up you would see 95% of the time the 14900 MAXOCWTFBBQ chip is getting stomped and dipping lower.

My problem is - I don't want to see plots and all that stuff. I don't want evidence, because I trust the people in my ocing community when we are sharing our results. By having someone pull crap like this it makes everyone look suspicious. I never show settings in any of my runs - I say it's ultra and it is ultra - and people believe me. But having crap like that put's a shadow over everything. Over what? Selling craptorlakes? Shilling for Intel? I don't get the goal.

 

[Akkedie]

I don't know if it's a big ask but do you perhaps have kingdom come installed?
I'm doing research on my next cpu upgrade and a thread came up about something called "amdip" which seems to be a bit of a controversial topic.
He runs trough a specific gate that's causing him a dip that supposedly intel doesn't suffer from.
If you have time could you perhaps go in the same spot in-game with your 9800x3d and see if you get the same dip?I was told the man in the video doesn't know how to tune amd cpu's and that its's probably flck related.
Thanks in advance.

The dip there is not to do with any asset streaming, it's actually due to that being an NPC area, and in this game the NPCs get much more development in terms of AI than most other OW games (life routines, etc. and they're always running; f.ex. even when fast travelling the simulation keeps running rather than being instanced - basically the world is alive, always; but when they're in your area there's even more happening with them, hence high CPU costs). The most CPU stressing areas is not that anyway, it's running through Rattay particularly around the market square, or the first major battle (because DX11 bottlenecks as an API wrt draw calls; engine is also very ST dependent, see any Crysis game).

 

[kapone32]

My problem is - I don't want to see plots and all that stuff. I don't want evidence, because I trust the people in my ocing community when we are sharing our results. By having someone pull crap like this it makes everyone look suspicious. I never show settings in any of my runs - I say it's ultra and it is ultra - and people believe me. But having crap like that put's a shadow over everything. Over what? Selling craptorlakes? Shilling for Intel? I don't get the goal.

The Goal? The more propaganda you spread the more clicks you get.

 

[kDude]

This is the problem, he claims he is running 6400c30 "MAX OC", and yet his factorio run is doing 4800 jedec ram. Obviously he is trying to sell his craptorlakes and his 1k$ a pop consults..

Spoiler

View attachment 371811

Here are my runs, Low settings like he is using in his video

Spoiler

And ultra settings which is how you are supposed to test

Spoiler

Am I seeing this right the low setting one you went into the gate with around 400 fps and dropped to 316 and the ultra setting went from 200 to 160?You said intel has the same drop,correct?I'm wondering how would a live frametime graph look going trough that gate.Would it be a frametime spike from the sudden drop or a smooth drop which wouldn't feel as a stutter.
Thanks for the videos bdw.
Edit:Oh it's the intel cpu not amd.

 

[JustBenching]

Am I seeing this right the low setting one you went into the gate with around 400 fps and dropped to 316 and the ultra setting went from 200 to 160?You said intel has the same drop?I'm wondering how would a live frametime graph look going trough that gate.
Thanks for the videos bdw.

The above videos are intel, it's a 12900k. At low settings it doesn't drop, at ultra settings it has a slight stutter. I have no idea how the 9800x 3d behaves, I haven't set up the PC yet, it's sitting in the corner looking at me.

 

[Dr. Dro]

The Goal? The more propaganda you spread the more clicks you get.

Since that video was posted, there was reportedly a second incident, also on an MSI motherboard

Updated: Two Ryzen 7 9800X3D CPUs burned out on X870 motherboards — vendor investigates the Ryzen burnout issues

It could be a user error or a bad batch of MSI motherboards with a defective socket.

www.tomshardware.com

The above videos are intel, it's a 12900k. At low settings it doesn't drop, at ultra settings it has a slight stutter. I have no idea how the 9800x 3d behaves, I haven't set up the PC yet, it's sitting in the corner looking at me.

Hahah I know the feel. Got a new liquid cooler for my 13900KS. Couldn't be bothered installing it yet. Think i'll just procrastinate, get a few extra upgrades and do it in one go.

 

[JustBenching]

Since that video was posted, there was reportedly a second incident, also on an MSI motherboard

Updated: Two Ryzen 7 9800X3D CPUs burned out on X870 motherboards — vendor investigates the Ryzen burnout issues

It could be a user error or a bad batch of MSI motherboards with a defective socket.

www.tomshardware.com

I think there is an another one on overclocknet (someone mentioned it somewhere) but I can't seem to find it. Maybe it was just a repeat of the same incident, no idea.

 

[kDude]

The above videos are intel, it's a 12900k. At low settings it doesn't drop, at ultra settings it has a slight stutter. I have no idea how the 9800x 3d behaves, I haven't set up the PC yet, it's sitting in the corner looking at me.

Yea that's exactly what I wanted to know then.Both intel and amd have the "dip/stutter" even if it's on the smaller side in that specific spot.
Real question is probably how high is the frametime spike when intel has it vs when amd has it when going trough that gate.
I'm keeping tabs on this thread incase you decided to make the same test with the 98003dx.I'm curious to see if you get a bigger more pronounced frametime spike or something in the range of the intel cpu.
Thanks for the testing.

 

[JustBenching]

Yea that's exactly what I wanted to know then.Both intel and amd have the "dip/stutter" even if it's on the smaller side in that specific spot.
Real question is probably how high is the frametime spike when intel has it vs when amd has it when going trough that gate.
I'm keeping tabs on this thread incase you decided to make the same test with the 98003dx.I'm curious to see if you get a bigger more pronounced frametime spike or something in the range of the intel cpu.
Thanks for the testing.

Is this to your satisfaction? It's the same run as in the video walking through the gate - 12900k low settings.

Spoiler

 

[kDude]

Is this to your satisfaction? It's the same run as in the video walking through the gate - 12900k low settings.

Spoiler

View attachment 371814

Is it supposed to look that spikey?average frametime of 2-3 ms with frequent ms spikes of up to 16 ms.I don't think i'd consider it playable if it was me lol. You did say it's on low setting so the crazy high fps could be at fault and this is on intel.

 

[JustBenching]

Is it supposed to look that spikey?average frametime of 2-3 ms with frequent ms spikes of up to 16 ms.I don't think i'd consider it playable if it was me lol. You did say it's on low setting so the crazy high fps could be at fault and this is on intel.

It is what it is, the game isn't the best optimized title.

 

[phanbuey]

Can cofirm -- it's one of my favorite games, but it's got a bit of that jank in it.

Played it on 8700k, 10850K, 3600x, 12600k and on 13700k -- they really stretched that engine, if you go above the 120-60 fps range it will remind you it's trying to do too much.

 

[Akkedie]

Troy Total War replay -> C:\Users\[user]\AppData\Roaming\The Creative Assembly\Troy_Steam\replays

test.replay

drive.google.com

Got to manually set settings to highest, most important Grass Extreme, Shadows Extreme, Unit Size Extreme. Obviously don't watch the whole thing, you can see fps go down a lot once the armies clash in the middle.

 

[kDude]

It is what it is, the game isn't the best optimized title.

You're right.
If I may ask another thing from you.When you're doing your side by side videos of the 12900k/98003dx could you enable the live frametime graph in msi afterburner?Feels like it's such an underrated setting that alot of content creators don't use when benchmarking games.Seeing just numbers going up and down sometimes doesn't reveal actual smoothness/un-smoothness unlike a frametime graph which can visually tell you a stutter has happened with a spike.(and how big or small it is)

 

[RootinTootinPootin]

Played it on 8700k, 10850K, 3600x, 12600k and on 13700k

You traitor..

 

[JustBenching]

You're right.
If I may ask another thing from you.When you're doing your side by side videos of the 12900k/98003dx could you enable the live frametime graph in msi afterburner?Feels like it's such an underrated setting that alot of content creators don't use when benchmarking games.Seeing just numbers going up and down sometimes doesn't reveal actual smoothness/un-smoothness unlike a frametime graph which can visually tell you a stutter has happened with a spike.

I already have finished with the 12900k footage

I might enable it just for the 9800x 3d - but im not using MSI AB.

A fix for 24h2 horrible Intel performance

24H2 CPU score drop in benchmarks (specifically TimeSpy)...

www.overclock.net

 

[kDude]

I already have finished with the 12900k footage

LOL no worries.

I might enable it just for the 9800x 3d - but im not using MSI AB.

Much appreciated.

 

[Akkedie]

Btw don't need comparison vs 12900k. Would be cool to see the result for just 9800X3D.

 

[kapone32]

A fix for 24h2 horrible Intel performance

24H2 CPU score drop in benchmarks (specifically TimeSpy)...

www.overclock.net

There are many issues for AMD users as well for 24H2.