Speculation: AMD's response to Intel's 8-core i9-9900K

Page 17 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

How will AMD respond to the release of Intel's 8-core processor?

  • Ride it out with the current line-up until 7nm in 2019

    Votes: 129 72.1%
  • Release Ryzen 7 2800X, using harvested chips based on the current version of the die

    Votes: 30 16.8%
  • Release Ryzen 7 2800X, based on a revision of the die, taking full advantage of the 12LP process

    Votes: 17 9.5%
  • Something else (specify below)

    Votes: 3 1.7%

  • Total voters
    179

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136
So, that is two codes out of...……………..?

I counted 3 if you took notice, that being said if those 14% were real then a 8700K should perform 25% better than a 2600X on average given the 10% frequency advantage, yet this average do not materialize elsewhere than in The Stilt s "results" :

https://www.computerbase.de/2018-04/amd-ryzen-2000-test/3/#diagramm-leistungsrating-anwendungen

https://www.hardware.fr/articles/975-17/indices-performance.html

9% at Computerbase and 11.3% at Hardware.fr, wich amount to the 8700K frequency advantage, in games it s 16% and 11.1% respectively, one more time perf is more frequency than throughput/Hz related...
 

coercitiv

Diamond Member
Jan 24, 2014
7,355
17,425
136
Can you even get an 8700K to stay at base frequency under load on it's own?
Sure you can, put an AVX load on it, maybe spice it up with some GPU load, it will eventually lower clocks to keep power at 95W IF this power target is configured as stock on the motherboard. (I doubt any Z board enforces stock TDP)

Anyway, my original post to which you replied with a hardly necessary explanation of how Intel TDP is defined was based on package power readings while running CB15 (heavy but realistic load) on CFL 6c/12t at 4.3Ghz. If I were to consider Prime 95 I would have said 110-120W instead. It so happens CB load pushes the CPU power usage in the same territory as TDP, but my point was average power consumption, as increasing core count and frequency would require process improvements to compensate. (otherwise the higher than 95W average power consumption would dictate higher TDP)

Since then I've been answered: we don't care about power anymore and related stock clocks either. I guess we'll end up comparing overclocked 2700X with overclocked 9900K and overclocked 2920X in MT loads and call it a day. (that is besides games where 9900K will be a god among puny chips)
 

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
This was all about using 15% from Stilts test as a means of comparing between processors based on core counts, like Epsilon did:



IMO a number derived only based off single core, opens source with lastest Intel compiler, is less representative IMO,than one using off the shelf software with SMT/HT enabled. But neither is really generally applicable.

IMO there are no blanket IPC numbers.

You really need a formula like

Clock Speed X Core Count X IPC of specific load(and accounting for HT/SMT).

So in the case of comparing a 8C 9900K to a 1920x, it depends heavily on what you are doing. Playing games, the 9900K will clearly be better, doing 3D rendering, the 1920x will have sizable lead.

I agree that there is no universal 'absolute IPC', it all depends on the software suite used. As abwx pointed out, you can actually find comparisons that show AMD having almost no IPC deficit, then others like techspot showing something in between, compared to Stilt's 'Intel favouring' results. Who is correct? They all are. Also, you are correct that there is ST IPC, and then there is MT IPC which has to take into account scaling of HT/SMT - the only problem is that this is a moving target as well, as HT/SMT improvements range from almost nothing (sometimes negative scaling even) to over 40% depending on the software.

FWIW, my point doesn't really change WR to the 9900X vs 1920X even if you change the 1.15x IPC advantage to something like 1.1x or even 1.05x

4.7 x 8 x 1.1 = 41.36
4.7 x 8 x 1.05 = 39.48
3.7 x 12 = 44.4

Let me put it another way - the 9900K would be closer in overall performance to a 1920X (or 2920X even) than it is to a 2700X. So I actually do agree that potentially pricing those chips close to the $500 mark could steal some of the 9900K thunder, but X399 motherboards won't be discounted anytime soon, unfortunately. Then again, when you are in this price range, do people care that you have to spend $300 on a motherboard?
 
  • Like
Reactions: Vattila

dlerious

Platinum Member
Mar 4, 2004
2,115
925
136
I agree that there is no universal 'absolute IPC', it all depends on the software suite used. As abwx pointed out, you can actually find comparisons that show AMD having almost no IPC deficit, then others like techspot showing something in between, compared to Stilt's 'Intel favouring' results. Who is correct? They all are. Also, you are correct that there is ST IPC, and then there is MT IPC which has to take into account scaling of HT/SMT - the only problem is that this is a moving target as well, as HT/SMT improvements range from almost nothing (sometimes negative scaling even) to over 40% depending on the software.

FWIW, my point doesn't really change WR to the 9900X vs 1920X even if you change the 1.15x IPC advantage to something like 1.1x or even 1.05x

4.7 x 8 x 1.1 = 41.36
4.7 x 8 x 1.05 = 39.48
3.7 x 12 = 44.4

Let me put it another way - the 9900K would be closer in overall performance to a 1920X (or 2920X even) than it is to a 2700X. So I actually do agree that potentially pricing those chips close to the $500 mark could steal some of the 9900K thunder, but X399 motherboards won't be discounted anytime soon, unfortunately. Then again, when you are in this price range, do people care that you have to spend $300 on a motherboard?
Prices have been released. The 2920X is $649 - I've seen the 1920X for $523. The 2950X is $899, 2970WX is $1299, and the 2990WX is $1799. 32-core launches 8/13 , 16-core 8/31, 12 and 24-core in October. It doesn't look like there's going to be an 8-core 2900X.
 

PeterScott

Platinum Member
Jul 7, 2017
2,605
1,540
136
I agree that there is no universal 'absolute IPC', it all depends on the software suite used. As abwx pointed out, you can actually find comparisons that show AMD having almost no IPC deficit, then others like techspot showing something in between, compared to Stilt's 'Intel favouring' results. Who is correct? They all are. Also, you are correct that there is ST IPC, and then there is MT IPC which has to take into account scaling of HT/SMT - the only problem is that this is a moving target as well, as HT/SMT improvements range from almost nothing (sometimes negative scaling even) to over 40% depending on the software.

FWIW, my point doesn't really change WR to the 9900X vs 1920X even if you change the 1.15x IPC advantage to something like 1.1x or even 1.05x

4.7 x 8 x 1.1 = 41.36
4.7 x 8 x 1.05 = 39.48
3.7 x 12 = 44.4

Let me put it another way - the 9900K would be closer in overall performance to a 1920X (or 2920X even) than it is to a 2700X. So I actually do agree that potentially pricing those chips close to the $500 mark could steal some of the 9900K thunder, but X399 motherboards won't be discounted anytime soon, unfortunately. Then again, when you are in this price range, do people care that you have to spend $300 on a motherboard?

But as mentioned I am skeptical of the rumored 4.7 all core clocking of 9900k. Let's just push them as far as reasonable.

Say you can OC 9900K to 5GHz, and 1920x to 4Ghz. Seems in the realm for both. So that yields.

5 x 8 = 40
vs
4 x 12 = 48

Before you get to IPC difference. If you have the load that is neutral or favors Ryzen then you have a case. But again, it is highly load dependent.

The $300 MB. Yeah, I thought there were $200 models, but it appears $300 is the floor. So it will definitely cost more.

Bottom line, I guess you need a load that is at least neutral, preferably favoring Ryzen, to make it an attractive alternative for someone considering a 9900K.

Probably more useful as marketing stunt that real sales movers. Like the 8086K trade in program.
 
  • Like
Reactions: coercitiv

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
But as mentioned I am skeptical of the rumored 4.7 all core clocking of 9900k. Let's just push them as far as reasonable.

Say you can OC 9900K to 5GHz, and 1920x to 4Ghz. Seems in the realm for both. So that yields.

5 x 8 = 40
vs
4 x 12 = 48

Before you get to IPC difference. If you have the load that is neutral or favors Ryzen then you have a case. But again, it is highly load dependent.

The $300 MB. Yeah, I thought there were $200 models, but it appears $300 is the floor. So it will definitely cost more.

Bottom line, I guess you need a load that is at least neutral, preferably favoring Ryzen, to make it an attractive alternative for someone considering a 9900K.

Probably more useful as marketing stunt that real sales movers. Like the 8086K trade in program.

Intel clocking a 9900K at 4.7GHz is no different to AMD clocking a 2700X at 4.05GHz all core turbo - they will both be right up against their respective clockspeed ceilings.

Sure, you can be sceptical, but CFL chips are capable of scaling to 4.7GHz at default voltages. Not every chip, mind you, but many do (including my 8700K) and certainly you would expect Intel to use the best dies on the 9900K, that along with a soldered IHS means I don't really have doubts that it's technically feasible. Whether these are the actual launch clocks, we'll find out soon.

Besides, I'm pretty sure most 9900K owners would be overclocking so I would expect 5GHz at a minimum considering the soldered IHS, so the stock speeds are somewhat of a moot point for many enthusiasts.

But you're right, how it compares against a 1920X/2920X will be highly load dependent. I think at the market it's targeted at (desktop users) the 9900K to me seems the better choice due to much higher ST and gaming performance whilst having competitive MT performance, plus lower platform costs.
 
Last edited:

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
Sure you can, put an AVX load on it, maybe spice it up with some GPU load, it will eventually lower clocks to keep power at 95W IF this power target is configured as stock on the motherboard. (I doubt any Z board enforces stock TDP)

Anyway, my original post to which you replied with a hardly necessary explanation of how Intel TDP is defined was based on package power readings while running CB15 (heavy but realistic load) on CFL 6c/12t at 4.3Ghz. If I were to consider Prime 95 I would have said 110-120W instead. It so happens CB load pushes the CPU power usage in the same territory as TDP, but my point was average power consumption, as increasing core count and frequency would require process improvements to compensate. (otherwise the higher than 95W average power consumption would dictate higher TDP)

Since then I've been answered: we don't care about power anymore and related stock clocks either. I guess we'll end up comparing overclocked 2700X with overclocked 9900K and overclocked 2920X in MT loads and call it a day. (that is besides games where 9900K will be a god among puny chips)

I think you pretty much answered your own questions - TDP limits aren't enforced on enthusiastic level motherboards by either AMD nor Intel, so it's rather meaningless to discuss hypotheticals when we all know under worst case loads the power consumption will far exceed the rated TDP.
If you want power efficiency then these chips aren't for you.

Bottom line - get a good cooler for a 9900K and you're set. Sure, it may consume 200W when overclocked to 5GHz, so what? A high end GPU consumes even more than that at stock and no one cares.
 

PeterScott

Platinum Member
Jul 7, 2017
2,605
1,540
136
Sure, you can be sceptical, but CFL chips are capable of scaling to 4.7GHz at default voltages. Not every chip, mind you, but many do (including my 8700K) and certainly you would expect Intel to use the best dies on the 9900K, that along with a soldered IHS means I don't really have doubts that it's technically feasible. Whether these are the actual launch clocks, we'll find out soon.

CFL is capable of 4.7GHz all core, but Intel tends to play that conservative and leave some headroom, for reliability. Let the Overclockers chase the extra. But out of the box they want the highest reliability they can get. Doing 4.3GHz means pretty much none will fail out at default clocks except is some extreme fluke situation.

As far as the soldered IHS, that is also a rumor. IMO when I saw those rumors together, it read like an enthusiast wish list. 8Cores with 5GHz Turbo, with 4.7GHz all core Turbo, and soldered IHS.

When I see rumors that look like a wish list, I can't help thinking there is some wishful thinking in there, or people having a laugh.

So I view those rumors with a LOT of skepticism.

Just like the rumored 8 Core CXX, giving 16 core 7nm Ryzen desktop parts in 2019, looks a lot like wishful thinking to me, so I am skeptical of that as well.
 
  • Like
Reactions: scannall

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
Prices have been released. The 2920X is $649 - I've seen the 1920X for $523. The 2950X is $899, 2970WX is $1299, and the 2990WX is $1799. 32-core launches 8/13 , 16-core 8/31, 12 and 24-core in October. It doesn't look like there's going to be an 8-core 2900X.

Reasonable prices all round, but with the exception of the 1920X, not really close enough to the 9900K to be truly disruptive. Still, the price of X399 mobos makes the 1920X a hard sell in a direct comparison to the 9900K, since we're looking at

$823 for a 1920X + X399 mobo ($300)
$600 for a 9900K + Z370/Z390 mobo ($150)

Quad channel memory could potentially be more expensive than dual channel as well.

I'm aware that X399 boards have additional features not found on Z370/Z390 so it's not exactly a direct apples to apples comparison, but that's what we have in front of us in terms of cost outlay.
 

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
CFL is capable of 4.7GHz all core, but Intel tends to play that conservative and leave some headroom, for reliability. Let the Overclockers chase the extra. But out of the box they want the highest reliability they can get. Doing 4.3GHz means pretty much none will fail out at default clocks except is some extreme fluke situation.

As far as the soldered IHS, that is also a rumor. IMO when I saw those rumors together, it read like an enthusiast wish list. 8Cores with 5GHz Turbo, with 4.7GHz all core Turbo, and soldered IHS.

When I see rumors that look like a wish list, I can't help thinking there is some wishful thinking in there, or people having a laugh.

So I view those rumors with a LOT of skepticism.

Just like the rumored 8 Core CXX, giving 16 core 7nm Ryzen desktop parts in 2019, looks a lot like wishful thinking to me, so I am skeptical of that as well.

I'm just merely speaking from what I believe is technically feasible, 4.7GHz is definitetly in the upper ranges of that, but still plausible.

I do see where your scepticism is coming from though, even though I believe the soldered IHS would be a great idea especially on a chip like the 9900K. Let's hope the rumours are true.
 

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
27,247
16,108
136
Reasonable prices all round, but with the exception of the 1920X, not really close enough to the 9900K to be truly disruptive. Still, the price of X399 mobos makes the 1920X a hard sell in a direct comparison to the 9900K, since we're looking at

$823 for a 1920X + X399 mobo ($300)
$600 for a 9900K + Z370/Z390 mobo ($150)

Quad channel memory could potentially be more expensive than dual channel as well.

I'm aware that X399 boards have additional features not found on Z370/Z390 so it's not exactly a direct apples to apples comparison, but that's what we have in front of us in terms of cost outlay.
Quad channel is just 4 sticks, either 4 singles, 2 double or 4 pack. They don't have to be matched. All my TR boxes are 2 sets of 2 sticks, and they all run 3466 or faster.

I think with quad channel, and all the other advantages, the 1920x would over be better, but it costs more.

Bottom line is we have more choices. Of course all this is dependent on the 9900k being $450, and I would not bet the farm on that. Good if it was.
 
  • Like
Reactions: Drazick and Vattila

PeterScott

Platinum Member
Jul 7, 2017
2,605
1,540
136
I'm just merely speaking from what I believe is technically feasible, 4.7GHz is definitetly in the upper ranges of that, but still plausible.

I do see where your scepticism is coming from though, even though I believe the soldered IHS would be a great idea especially on a chip like the 9900K. Let's hope the rumours are true.

I too think it would be a great idea. Like I said,it reads almost like a wish list.

I assume the i7-9700K (8c8t) would also get that soldered IHS if the 9900K gets it, and maybe similar clock speeds, which could help it have a performance delta over 8700K (6c12t).

I wish they would announce this thing. I am so curious to see what it gets. Then I will be so dang curious to see what 7nm Ryzen gets. :D
 
  • Like
Reactions: epsilon84

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
Quad channel is just 4 sticks, either 4 singles, 2 double or 4 pack. They don't have to be matched. All my TR boxes are 2 sets of 2 sticks, and they all run 3466 or faster.

I think with quad channel, and all the other advantages, the 1920x would over be better, but it costs more.

Bottom line is we have more choices. Of course all this is dependent on the 9900k being $450, and I would not bet the farm on that. Good if it was.
Yeah I understand that,
Quad channel is just 4 sticks, either 4 singles, 2 double or 4 pack. They don't have to be matched. All my TR boxes are 2 sets of 2 sticks, and they all run 3466 or faster.

I think with quad channel, and all the other advantages, the 1920x would over be better, but it costs more.

Bottom line is we have more choices. Of course all this is dependent on the 9900k being $450, and I would not bet the farm on that. Good if it was.

I see your point, I was talking about getting matched pairs, for example:
https://www.newegg.com/Product/Prod...2gb_trident_X_2_x_16gb-_-20-232-560-_-Product
https://www.newegg.com/Product/Prod...channel_32gb_trident_X-_-20-232-650-_-Product

The 1920X would have a small MT advantage but significant ST deficit, as well as gaming. If you run MT apps all day (which, to be fair, a TR user very well might) then sure, it could be the better chip. For any other mixed use aka general desktop usage I would prefer the 9900K.
 

coercitiv

Diamond Member
Jan 24, 2014
7,355
17,425
136
I think you pretty much answered your own questions - TDP limits aren't enforced on enthusiastic level motherboards by either AMD nor Intel, so it's rather meaningless to discuss hypotheticals when we all know under worst case loads the power consumption will far exceed the rated TDP. If you want power efficiency then these chips aren't for you.

Bottom line - get a good cooler for a 9900K and you're set. Sure, it may consume 200W when overclocked to 5GHz, so what? A high end GPU consumes even more than that at stock and no one cares.
I had no issue with using overclocked 9900K numbers, I had no issue with non-stock power numbers, what I did have an issue with was using stock clocks for 1920X in a comparison with obviously non-stock clocks for 9900K.

You said it yourself, enthusiasts won't bother with power draw, so why do you insist on using 3.7Ghz for 1920X?
 

ub4ty

Senior member
Jun 21, 2017
749
898
96
Reasonable prices all round, but with the exception of the 1920X, not really close enough to the 9900K to be truly disruptive. Still, the price of X399 mobos makes the 1920X a hard sell in a direct comparison to the 9900K, since we're looking at

$823 for a 1920X + X399 mobo ($300)
$600 for a 9900K + Z370/Z390 mobo ($150)

Quad channel memory could potentially be more expensive than dual channel as well.

I'm aware that X399 boards have additional features not found on Z370/Z390 so it's not exactly a direct apples to apples comparison, but that's what we have in front of us in terms of cost outlay.

https://slickdeals.net/f/11909115-a...op-processor-yd195xa8aewof-599-shipped-amazon
1950x dropped down to $599 for some time today
https://old.reddit.com/r/buildapcsales/comments/957e3n/cpu_amd_ryzen_threadripper_1920x_24900/
1920x dropped down to $249

HAHA-GIF.gif

Momma Su straight up spitting on Intel's grave. The absolute state....
Rest in pieces and enjoy the great times ahead !

The next couple of years of hardware is going to be like a freight train w/ no brakes
 
  • Like
Reactions: Drazick

Zucker2k

Golden Member
Feb 15, 2006
1,810
1,159
136
I counted 3 if you took notice, that being said if those 14% were real then a 8700K should perform 25% better than a 2600X on average given the 10% frequency advantage, yet this average do not materialize elsewhere than in The Stilt s "results" :

https://www.computerbase.de/2018-04/amd-ryzen-2000-test/3/#diagramm-leistungsrating-anwendungen

https://www.hardware.fr/articles/975-17/indices-performance.html

9% at Computerbase and 11.3% at Hardware.fr, wich amount to the 8700K frequency advantage, in games it s 16% and 11.1% respectively, one more time perf is more frequency than throughput/Hz related...
Yes, but those results are derived from less than 10 applications (I believe 9, and 7 respectively). They are also highly threaded, and therefore plays into AMD's favour. I don't even know what methodology they used, but if you're going to test for IPC, you must test single core and multi-core separately. In the case of the latter, you must divide the total score by the number of threads. The Stilt's work and range of software covered is far superior to those benchmarks you referenced.

On typical desktop applications (gaming, photo processing, music encoding, even video), intel's ipc advantage plus superior clocks pushes Intel clearly ahead. You have to try very hard to find a program that'll max all cores on a desktop processor for an extended period of time. It's just unrealistic.
I agree that there is no universal 'absolute IPC', it all depends on the software suite used. As abwx pointed out, you can actually find comparisons that show AMD having almost no IPC deficit, then others like techspot showing something in between, compared to Stilt's 'Intel favouring' results. Who is correct? They all are. Also, you are correct that there is ST IPC, and then there is MT IPC which has to take into account scaling of HT/SMT - the only problem is that this is a moving target as well, as HT/SMT improvements range from almost nothing (sometimes negative scaling even) to over 40% depending on the software.

FWIW, my point doesn't really change WR to the 9900X vs 1920X even if you change the 1.15x IPC advantage to something like 1.1x or even 1.05x

4.7 x 8 x 1.1 = 41.36
4.7 x 8 x 1.05 = 39.48
3.7 x 12 = 44.4

Let me put it another way - the 9900K would be closer in overall performance to a 1920X (or 2920X even) than it is to a 2700X. So I actually do agree that potentially pricing those chips close to the $500 mark could steal some of the 9900K thunder, but X399 motherboards won't be discounted anytime soon, unfortunately. Then again, when you are in this price range, do people care that you have to spend $300 on a motherboard?
But there is. There are even standard industry benchmarks. On the other points, read my reply to ABWX above.
 

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
I had no issue with using overclocked 9900K numbers, I had no issue with non-stock power numbers, what I did have an issue with was using stock clocks for 1920X in a comparison with obviously non-stock clocks for 9900K.

You said it yourself, enthusiasts won't bother with power draw, so why do you insist on using 3.7Ghz for 1920X?

The rumoured STOCK clocks for the 9900K are 4.7GHz all core turbo / 5.0GHz single core turbo. I don't know how much clearer I can make that.

https://wccftech.com/intel-9th-gen-core-i9-9900k-core-x-refresh-cpu-family-october-launch/

If you have another source that contradicts these leaked specs, then by all means, show us. But don't accuse me of 'overclocking' the 9900K when I'm clearly going by the rumoured specs.

If I were to use overclocked numbers for both chips it would be a 5GHz 9900K and a 4GHz 1920X - hardly changes the equation from 4.7 vs 3.7 does it?!
 
Last edited:
  • Like
Reactions: french toast

coercitiv

Diamond Member
Jan 24, 2014
7,355
17,425
136
If you have another source that contradicts these leaked specs, then by all means, show us. But don't accuse me of 'overclocking' the 9900K when I'm clearly going by the rumoured specs.
Running the chip beyond power specs is overclocking. If you want to be extremely specific and claim overclocking only happens when CPU runs beyond official multiplier for the number of cores, I would ask you what better term you have for non-stock power settings, especially considering power limits & temps are now being used to define stock clock behavior in some CPUs from both Intel and AMD as well as GPUs from both Nvidia and AMD.

My desktop CPU is a i7 8700 currently running without a 65W TDP limit. If you think that's not overclocking, please tell me what wording we can use to define this CPU running beyond manufacturer specs.
 

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
Running the chip beyond power specs is overclocking. If you want to be extremely specific and claim overclocking only happens when CPU runs beyond official multiplier for the number of cores, I would ask you what better term you have for non-stock power settings, especially considering power limits & temps are now being used to define stock clock behavior in some CPUs from both Intel and AMD as well as GPUs from both Nvidia and AMD.

My desktop CPU is a i7 8700 currently running without a 65W TDP limit. If you think that's not overclocking, please tell me what wording we can use to define this CPU running beyond manufacturer specs.
Turbo boost != overclocking otherwise by definition EVERY SINGLE CPU with turbo boost clocks is 'overclocked' or 'pre overclocked'

Overclocking to me means going into the BIOS and manually adjusting the multiplier higher than the top turbo multiplier. Clearly you have a different idea of what overclocking means and if we can't agree on what constitutes overclocking then this is a pointless discussion
 

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136
but if you're going to test for IPC, you must test single core and multi-core separately.

That doesnt make sense in The Stilt methodology because he use ST scores of softs that are exclusively used in MT in normal usage, i mean what is the use of ST scores of say 3DPM, Cinebench, X264, X265 FI.?

Is one going to render a scene using a single thread, or encode this way..?

This is only intermediary results that have no relevance with the application usage, but still, this is considered as representative of real world perf, lol...

If ST perf is to be accounted it must be measured with softs that are actually ST bound, like some parts of Solidworks or Catia 3D.
 
Last edited:

JoeRambo

Golden Member
Jun 13, 2013
1,814
2,105
136
Isn't this redefined "overclocking" even more irrelevant than "TDP"? We are talking about what is going to be fastest desktop CPU ( for general use, gaming and workloads that benefit from fast ST and MT into dozen of threads ). I understand some guys here are on quest to pour smelly substances on Intel's watermill and some are here to downplay the revolutionary value and performance AMD provides with Threadrippers for DC and rendering workloads. What I don't get is the retarded ways they do so, like:

1) Focusing on TDP, when none of vendors and mb ODMs are even following it and both are happy to run north of 120-130w with CB like loads, 95w TDP/CDP be damned.
2) Providing crazy calculations about 8C16T CFL beating certain amount of Threadripper cores, that lead to certain AMD fans splitting heirs and attacking the work of well respected forum members. Well, here is the news for you, for general desktop use, Intel is going to be ~30% faster, and for non-casual rendering/DC work You are probably already running TR or even multiple of them like Mark does.

I have 8700K at home, but probably going to buy 9900K, because it has 16MB of L3 and I hope will provided 300-400Mhz more over my current 4.8Ghz clock. Yeah, that is how much i value desktop performance, and that is knowing that run compiles, JVMs and sometimes multiple linux/android VMs.
 

StinkyPinky

Diamond Member
Jul 6, 2002
6,973
1,276
126
https://slickdeals.net/f/11909115-a...op-processor-yd195xa8aewof-599-shipped-amazon
1950x dropped down to $599 for some time today
https://old.reddit.com/r/buildapcsales/comments/957e3n/cpu_amd_ryzen_threadripper_1920x_24900/
1920x dropped down to $249

HAHA-GIF.gif



Momma Su straight up spitting on Intel's grave. The absolute state....
Rest in pieces and enjoy the great times ahead !

The next couple of years of hardware is going to be like a freight train w/ no brakes

Damn $249 for a 1920x?! Has to be a mistake.
 
  • Like
Reactions: Drazick