Is it really over for Intel®?

[Koenig168]

...

Theres also been a recent price drop on 265K/KF chip, so the cost to entry for this platform is now greatly reduced, coupled together with a good Mini ITX board like Asrock Z890i Nova won't even break $900 SGD.

...

AMD has the best gaming CPU and Intel is now the budget option

How the mighty have fallen ....

 

[alauz]

How the turntables, Intel now I wouldn't even recommend for production, that one just buy 9950 or 9950X3D if you want work & play at max level.

 

[ragnarok95]

How the turntables, Intel now I wouldn't even recommend for production, that one just buy 9950 or 9950X3D if you want work & play at max level.

Actually 285K is good for work, it's on par with 9950X (win some, lose some) at a lower pricing. But, if you throw gaming into the equation, that is where AMD is way better.

 

[elmariachi]

I struggle to also argue for Intel at this point but users like me and matique do not find a point to upgrade even to AMDs latest X3d CPUs. The performance difference between a fully tuned 13/14th Gen is probably on par or slightly below AMDs latest for gaming at 1440p which most people's use case is. Unless you have money to throw or you die die need efficiency, then it makes sense to go to AMD if you already have a 13/14th Gen. If you're not these group of users, AMD hands down as nothing beats the ease of out of the box performance which is pretty incredible.

 

[fkchon]

Nova Lake next big upgrade for those on fully tuned 13th/14th then?

 

[Encrypted11]

No indication is Nova Lake is led by Haifa (Israel Design Center), unlike Lunar Lake and Panther Lake.

If the past is any indication the design win by a lab is a tick+tock pair. I think Nova Lake is an Oregon project.

 

[elmariachi]

Nova Lake next big upgrade for those on fully tuned 13th/14th then?

Doubt Intel will make a comeback soon if they can't get their own foundaries in order. Seems like TSMC isn't really working out for them based on their design. Will go down AMD if next platform is pathetic like Intel. I only hope the stutter issues with the X3D variants is fixed when the cache is full and emptying.

 

[Phen8210]

I don't think it will be fixed since after exceeding the cache the rest of workload becomes dependent on the rest of the chip. this is the weakness of being heavily dependent on the 3D cache. If the system is without tuned memory then the dips will be pretty large.

If AMD wants to solve this issue then they really need to have high memory speeds running at 1:1 which i doubt will happen anytime soon.

 

[elmariachi]

I don't think it will be fixed since after exceeding the cache the rest of workload becomes dependent on the rest of the chip. this is the weakness of being heavily dependent on the 3D cache. If the system is without tuned memory then the dips will be pretty large.

If AMD wants to solve this issue then they really need to have high memory speeds running at 1:1 which i doubt will happen anytime soon.

That is the thing. The games I play runs very smoothly on the 9800X3D but after a while, there will be this massive 1s pause then suddenly the FPS shoots back up. I mean it is not a big deal but if you are in the middle of the gunfight in multiplayer, you are dead. The Intel has variances in fps drops but it does not hiccup like the way the AMD does. It is much smoother in the way it does so. The 9800X3D Frame times also shoots up like mad during that 1s then it quickly ramps up the FPS and Frame times go back down to normal levels. But this only happens 1 or 2 times in a game of around 12 mins so not too bad.

But I also heard these stutters are caused by RTSS and Afterburner in the recent updates.

 

[fkchon]

I don't think it will be fixed since after exceeding the cache the rest of workload becomes dependent on the rest of the chip. this is the weakness of being heavily dependent on the 3D cache. If the system is without tuned memory then the dips will be pretty large.

If AMD wants to solve this issue then they really need to have high memory speeds running at 1:1 which i doubt will happen anytime soon.

Could happen next gen since the memory controller is on the IO Die and that hasn't been updated for generations, they are doing a die shrink of the IO Die with Zen 6.

 

[Yongkit]

That is the thing. The games I play runs very smoothly on the 9800X3D but after a while, there will be this massive 1s pause then suddenly the FPS shoots back up. I mean it is not a big deal but if you are in the middle of the gunfight in multiplayer, you are dead. The Intel has variances in fps drops but it does not hiccup like the way the AMD does. It is much smoother in the way it does so. The 9800X3D Frame times also shoots up like mad during that 1s then it quickly ramps up the FPS and Frame times go back down to normal levels. But this only happens 1 or 2 times in a game of around 12 mins so not too bad.

But I also heard these stutters are caused by RTSS and Afterburner in the recent updates.

i think afterburner does not fully support 9800x3d, i noticed at least when i using it does not give me the correct reading for 1% low

 

[matique]

My next CPU will coincide with next GPU/monitor upgrade.

Will be looking for a 5K2K 240hz 34-38" OLED monitor to upgrade to first. This would need a lot of graphical power, which i expect the next 6090 to deliver as it would be a node shrink. As games start being more reliant on CPU perf again due to dlss and such, I will see which cpu and platform to go to then.

If i'm going AMD, hopefully it's AM6 so I'll be able to upgrade CPUs in the future, without changing mobos. If it's intel, i'll only go with them if they solve die latency as well as gaming perf.

I hope intel goes back to intel foundries for their chips, as they have more overhead factored in vs CPU running on limits out of the box from TSMC. 285K can't even OC much on main clock, only a bit more for the D2D & NGU.

 

[Phen8210]

i think afterburner does not fully support 9800x3d, i noticed at least when i using it does not give me the correct reading for 1% low

what reading does it give?

 

[elmariachi]

i think afterburner does not fully support 9800x3d, i noticed at least when i using it does not give me the correct reading for 1% low

4.6.6 Beta 5 is what you need to be running. I've had no issues with readings. The issue is that RTSS when I was running it made stutters worse. When I disabled it, it was much better.

 

[Yongkit]

what reading does it give?

1fps~5

 

[Yongkit]

4.6.6 Beta 5 is what you need to be running. I've had no issues with readings. The issue is that RTSS when I was running it made stutters worse. When I disabled it, it was much better.

Yes using 4.6 6 beta 5

 

[fkchon]

GSkill 8800 Kit is here, the black mirror finish looks more premium than the regular DDR5 GSkill Royals.

9600 boots easily on the Z890 Apex, damn thats alot of read bandwidth. Not sure if this will ever be usable since it has 80ns, might be able to tune it down to like 77-78 but thats it. Gear 2 is still the way to go.

I may need to copy these settings on the Teamgroup kit and use it on Taichi OCF, it boots but it wasn't stable maybe because the board preset settings were a bit too tight since theres no default profile for 9600 G4.

On a side note, just realized Easter this year is towards end of April. So no Mora purchase until then.

 

[Phen8210]

1fps~5

View attachment 53628

try starting and stopping the benchmark for avg and 1% lows manually, you must have it at on mode all the time.

 

[fkchon]

These GSkill 8800 CUDIMM are no good. Can't do XMP 8800 G2, immediately spits out error within first cycle of TM5. The non CUDIMM 8200 C36 Biwin rams can do 8800 with tighter timings and completes the test with no errors.

Will do more testing...

 

[elmariachi]

These GSkill 8800 CUDIMM are no good. Can't do XMP 8800 G2, immediately spits out error within first cycle of TM5. The non CUDIMM 8200 C36 Biwin rams can do 8800 with tighter timings and completes the test with no errors.

Will do more testing...

Try swapping the sticks around the slots and try again.
Always use the lower serial number closer to the CPU then the other one next to it.

If still have errors, likely you need to tune IMC Voltage and TX Voltage to get it to work.
Also, try reducing your SA down abit as I heard the default SA for 285K is pretty high.