[HN Gopher] Intel Rocket Lake (14nm) Review: Core i9-11900K, Cor...
___________________________________________________________________
 
Intel Rocket Lake (14nm) Review: Core i9-11900K, Core i7-11700K,
Core i5-11600K
 
Author : pantalaimon
Score  : 33 points
Date   : 2021-03-30 16:04 UTC (6 hours ago)
 
web link (www.anandtech.com)
w3m dump (www.anandtech.com)
 
| kllrnohj wrote:
| How on earth does Intel think 9 skus that are all 6 cores / 12
| threads over a difference of $100 makes any sense? 9 products
| from $157 to $262 that are all not really that different.
| 
| And 5 different products all called 11700 and 5 different
| products called 11900 for a total of 10 products with 8c / 16t
| spanning from $298 to $539?
 
  | bryanlarsen wrote:
  | The main market for Intel is not the consumer, it's the HP's
  | and Dell's of the world.
 
    | kllrnohj wrote:
    | Even still those are selling to consumers. How do you expect
    | an HP or a Dell to figure out how to sell a user on an
    | i5-11500 instead of the slower & more expensive i5-11600T?
    | 
    | Why isn't the i5 lineup just 11600 = the 125w part, 11500 =
    | 65w part, and 11400 = 35w part? Is there _really_ enough gap
    | in product stacks to justify a different 11400T vs. 11500T at
    | a $10 difference?
 
      | leetcrew wrote:
      | it doesn't really make sense for dell to offer a choice
      | between T and non-T for the same model. the only reason to
      | use a T part is to stay under a power/TDP budget.
 
      | andrewf wrote:
      | Is it possible that this proliferation of SKUs is how Intel
      | gives each of HP, Dell, Lenovo, Acer, etc exactly what they
      | are asking for? (I presume, if you buy enough Intel CPUs,
      | you have purchasing managers who are constantly chatting
      | back and forth with sales engineer types at Intel)
 
      | vbezhenar wrote:
      | It's about silicone lottery. They could ignore best chips
      | and sell them along bad chips at the same price. Or they
      | could market best chips as another product with a higher
      | price. And that's the whole spectrum.
 
        | kllrnohj wrote:
        | Sure, but we're talking a $10, or ~5%, difference. Either
        | these bin identically (since Intel is charging an
        | ~identical amount for them), or one of these will be
        | permanently out of stock making it functionally not exist
        | in the stack anyway. The actual profit difference here
        | will be even less as simply having that extra sku costs
        | overhead. So what practically changes if the 11500T just
        | didn't exist?
        | 
        | i7-11700T vs. i7-11900T would be silicon lottery in play,
        | that's $323 vs. $439 for a +100mhz clock bump. Although
        | then there's the question of "what is this 35w desktop
        | CPU market that will pay for a $439 CPU?"
 
  | pantalaimon wrote:
  | What boggles my mind are the inconsitencies. E.g. a i9 10900F
  | is clocked lower than a i9 10850K
  | 
  | Wouldn't you think higher number = faster, especially when core
  | count is the same?
 
    | josalhor wrote:
    | I have been incredibly skeptical of frequency numbers. Case
    | in point, the AMD FX-9590 had a 4.7 GHz base an 5.0 GHz boost
    | back in 2013 [0]; and the Ryzen 9 5950x has a 3.4 GHz base an
    | 4.9 GHz boost [1].
    | 
    | To me, the frequency has become a "blob". It means nothing.
    | If anything, I see more frequency and I think more power draw
    | and less stability.
    | 
    | Real performance is a function of IPC and frequency and that
    | is the only thing that really matters.
    | 
    | [0]: https://www.amd.com/es/products/cpu/fx-9590
    | 
    | [1]: https://www.amd.com/en/products/cpu/amd-ryzen-9-5950x
 
      | leetcrew wrote:
      | none of those numbers fully describe the performance of a
      | CPU, but idk why you would be "skeptical" of frequency
      | numbers in particular. IPC itself is somewhat arbitrary. in
      | the real world, IPC constantly varies with the mix of
      | instructions, cache-friendliness of the program/data,
      | memory performance, etc.
 
        | josalhor wrote:
        | > none of those numbers fully describe the performance of
        | a CPU, but idk why you would be "skeptical" of frequency
        | numbers in particular
        | 
        | Because I have seen quite a few consumers get guided too
        | much by the "more frequency, more better" motto.
        | 
        | Moreover, it doesn't stop there. I have seen papers where
        | the only thing cited is the frequency of the processors,
        | and no other reference to the
        | manufacturer/architecture/version of the processors used.
        | 
        | > IPC itself is somewhat arbitrary. in the real world,
        | IPC constantly varies with the mix of instructions,
        | cache-friendliness of the program/data, memory
        | performance, etc.
        | 
        | Absolutely, I would also be skeptical of an "IPC" metric.
        | Performance is the only thing that counts.
 
      | tenacious_tuna wrote:
      | > To me, the frequency has become a "blob". It means
      | nothing.
      | 
      | That seems reasonable across microarch versions, but
      | presumably is totally reasonable within a product-release-
      | cycle? i.e. to compare the FX-9590 to an FX-8350 based off
      | clock speed. If it's the same core count and microarch,
      | presumably clock is about the only difference, aside from
      | other features like PCIE lanes, cache, etc.
 
        | josalhor wrote:
        | > That seems reasonable across microarch versions, but
        | presumably is totally reasonable within a product-
        | release-cycle
        | 
        | Sure, but the performance doesn't scale linearly with the
        | frequency. So how can we account for that? Also, the i9
        | 10900F and i9-10850K have the same turbo. How long can
        | they sustain in for? On how many cores? What frequency is
        | left afterward? This gets even weirder with more recent
        | technologies that push the frequency higher depending on
        | power delivery stability and thermal constraints.
        | 
        | I agree; the frequency is indicative of performance
        | within a product lifecycle. But I find it too meaningless
        | to give an approximate delta on the performance. And this
        | assumes the process you are running is not too much
        | memory bottlenecked or I/O bound and this delta in
        | performance can be measured. (And the setup can sustain
        | higher thermals).
        | 
        | While not perfect, I prefer to look up benchmarks that
        | resemble my workload and call it a day. Processors are
        | too complicated for me to try to compare with simple
        | numbers.
 
    | belval wrote:
    | This is not really new, it's because of the letter at the end
    | of the product name. Traditionally Ks are power hungry
    | overclockable chips while F are more energy-conscious.
    | 
    | EDIT: Never mind I just confirmed your point. T are the
    | energy-conscious ones, not sure what F are. This is all very
    | confusing.
 
      | leetcrew wrote:
      | F means it lacks an IGP. in a certain sense you were right;
      | the F variant sacrifices frequency for lower TDP vs a K
      | variant. but there are also KF parts that are unlocked and
      | also lack an IGP. and of course there's a different system
      | entirely for laptop parts.
 
| AmVess wrote:
| Decent IPC increase, but really good liquid cooling is required
| in order to get it. Prebuilts like Dell will likely use tiny
| coolers that will limit the performance of these parts.
| 
| Apple's M1 illustrates how far behind Intel is in everything, and
| AMD's upcoming Zen 4 is supposed to have a staggering 20%
| increase in IPC.
| 
| It is a good thing Intel has massive market inertia, because the
| next few years are going to be rough for them from a tech
| standpoint.
 
  | lmedinas wrote:
  | Actually to me this generation seems like its even a regression
  | compared to the previous due to the reduction of cores and
  | increase of Power consumption.
  | 
  | Does look good for Intel :/
 
___________________________________________________________________
(page generated 2021-03-30 23:02 UTC)