Intel's new 14900K rated in horsepower

D

dada_dave

Elite Member
Posts
2,164
Reaction score
2,151
Screen Shot 2023-10-17 at 9.56.48 AM.png


I know it's often discussed how "watts don't matter for desktop users". But there is a limit to that I would think ... especially if you live in an area where the price of energy is not so low! In fairness, if you read the review (link below) this is partially if not wholly on the motherboard manufacturers but still.

Screen Shot 2023-10-17 at 9.58.22 AM.png


Full review:

www.anandtech.com

Intel Core i9-14900K, Core i7-14700K and Core i5-14600K Review: Raptor Lake Refreshed

www.anandtech.com www.anandtech.com
 
Jimmyjames

Jimmyjames

Site Champ
Posts
675
Reaction score
763
Nice headline. I was gonna go with “the worst Lake since Camp Crystal”.

Only skimmed the review, but it seems like a poor release? Hardly any performance increase with no meaningful changes to either P or E cores, but just more power. Is that a fair summary?

Highest GB 6 score I saw was -3300.

Iirc the next release is mobile focused, so no increases in top end performance for a while.
 
D

dada_dave

Elite Member
Posts
2,164
Reaction score
2,151
Jimmyjames said:
Nice headline. I was gonna go with “the worst Lake since Camp Crystal”.

Only skimmed the review, but it seems like a poor release? Hardly any performance increase with no meaningful changes to either P or E cores, but just more power. Is that a fair summary?
Click to expand...

As far as I can tell, yes.
 
Aaronage

Aaronage

Power User
Posts
144
Reaction score
213
The idea that power consumption/efficiency doesn’t matter on the desktop has always been bollocks.
It mattered with NetBurst, it mattered with Bulldozer, and it matters today with *Cove/Lake 🔥🥵
 
D

dada_dave

Elite Member
Posts
2,164
Reaction score
2,151
Aaronage said:
The idea that power consumption/efficiency doesn’t matter on the desktop has always been bollocks.
It mattered with NetBurst, it mattered with Bulldozer, and it matters today with *Cove/Lake 🔥🥵
Click to expand...
I can understand the criticism that Apple should push the power curves on their desktop products harder and I do agree that it matters *less* on that product. But some of these products from Nvidia/Intel/motherboards are getting excessive even by enthusiast standards I would think. I mean I would also add that if your argument is that perf/$ on the desktop matters more than perf/W, you should still have to calculate $ in terms of the expected energy costs of running the computer in your area. For some, that may make little difference. For others? Quite a bit. Especially with the ridiculous consumptions we're starting to see. Never mind the noise factor ...
 
rdrr

rdrr

Elite Member
Posts
1,229
Reaction score
2,056
The energy consumption worry is not just impacting personal use. A lot of research computing it putting strain on local power companies. Our reseach department is looking to replace the aging HPC and the power consumption is limiting what they can get. The local power company is unwilling to give the building more power, and don't even talk to me about the costs of HPC in the Cloud.
 
D

dada_dave

Elite Member
Posts
2,164
Reaction score
2,151
rdrr said:
The energy consumption worry is not just impacting personal use. A lot of research computing it putting strain on local power companies. Our reseach department is looking to replace the aging HPC and the power consumption is limiting what they can get. The local power company is unwilling to give the building more power, and don't even talk to me about the costs of HPC in the Cloud.
Click to expand...
Very true, power consumption "not mattering" was only ever a semi-reasonable talking point for desktop gaming PCs/home workstations. For HPC/cloud/server? power consumption matters a lot again.
 
Last edited:
T

theorist9

Site Champ
Posts
613
Reaction score
563
For desktop use, I think noise is one of the most irritating issues created by high power consumption. This can be mitigated with a large and expensive case—but then size and expense can also be an issue for desktop users.

I was going to add that, if you're using it in an air-conditioned room, that you have to add, to the utility cost, the cost of running the AC to remove the added thermal energy. But I did the calculation, and that's marginal--about an added 3% if your AC has an EER (Energy Efficiency Ratio) of 10. Confusingly, the EER isn't (watts removed)/(watts to run AC), it's (BTU/hr removed)/(watts to run AC), and 1 BTU/hr = 0.293 W. Thus, if the AC's EER is 10, you need to put an additional 1 W into running your AC for every additional 35 W your computer consumes.
 
Last edited:
T

theorist9

Site Champ
Posts
613
Reaction score
563
Yoused said:
Intel TDP: now you understand why the names end in "Lake".
Click to expand...
For some reason this reminds me of a triathlon I did where the swim portion was in Lake Anna, VA, an artifical lake created to cool the North Anna Nuclear Generating Station. I found it pleasantly warm.
 
D

dada_dave

Elite Member
Posts
2,164
Reaction score
2,151
Jimmyjames

Jimmyjames

Site Champ
Posts
675
Reaction score
763
Aaronage

Aaronage

Power User
Posts
144
Reaction score
213
It’s insane to me that Intel manages to sell ANY of these things to the DIY market. You have to be a total masochist to choose that dead-end socket with incredibly inefficient CPUs over AM5 and Ryzen 7000.

At least when AMD dropped the ball with Bulldozer they priced them accordingly. Raptor Refresh isn’t even good value 😂
 
Aaronage

Aaronage

Power User
Posts
144
Reaction score
213
On this topic, Emerald Rapids will supposedly have a cTDP up to 922W (or 1.23HP lol)
www.hardwaretimes.com

Intel's Xeon Emerald Rapids CPUs Leak Out: 448MB of Cache a TDP of up to 922W | Hardware Times

Intel’s next-gen Xeon “Emerald Rapids” processors have a release date: The 14th of December alongside the “Meteor Lake” Core Ultra family. Featuring denser logic and more cache, these chips refresh the 4th Gen Sapphire Rapids lineup on the same node and architecture. The Xeon Platinum 8592+ sits...
www.hardwaretimes.com www.hardwaretimes.com

I’m sure that 64-core waste of silicon will fare well against Amazon’s new 96-core Neoverse V2 chip 😜
 
D

dada_dave

Elite Member
Posts
2,164
Reaction score
2,151
Aaronage said:
It’s insane to me that Intel manages to sell ANY of these things to the DIY market. You have to be a total masochist to choose that dead-end socket with incredibly inefficient CPUs over AM5 and Ryzen 7000.

At least when AMD dropped the ball with Bulldozer they priced them accordingly. Raptor Refresh isn’t even good value 😂
Click to expand...
It’s insane that they lie so badly about “TDP” and that DIYers don’t have pitchforks in front of Intel’s offices. Peak power is over 3x TDP for the 14900X! That’s … well … insane which we’ve collectively used 3 times now. Okay, sure AMD low balls their TDP on their consumer parts, but at least it’s in the same ballpark.
 
Yoused

Yoused

up
Posts
5,624
Reaction score
8,944
Location
knee deep in the road apples of the 4 horsemen
Aaronage said:
On this topic, Emerald Rapids will supposedly have a cTDP up to 922W (or 1.23HP lol)
www.hardwaretimes.com

Intel's Xeon Emerald Rapids CPUs Leak Out: 448MB of Cache a TDP of up to 922W | Hardware Times

Intel’s next-gen Xeon “Emerald Rapids” processors have a release date: The 14th of December alongside the “Meteor Lake” Core Ultra family. Featuring denser logic and more cache, these chips refresh the 4th Gen Sapphire Rapids lineup on the same node and architecture. The Xeon Platinum 8592+ sits...
www.hardwaretimes.com www.hardwaretimes.com

I’m sure that 64-core waste of silicon will fare well against Amazon’s new 96-core Neoverse V2 chip 😜
Click to expand...
I looked up Neoverse on the wiki thingie, and saw where it said the V2 has 15 execution ports. What exactly does that mean?
 

Similar threads

Andropov
Intel Raptor Lake
2 3
Replies
53
Views
2K
Cmaier
Cmaier
T
Request for Adobe Acrobat Pro benchmarking
Replies
9
Views
564
dada_dave
D
Top Bottom
1 2