I9 9900k Official Reviews from Anandtech, Tomshardware. Add your own links to others !

[moinmoin]

Based on my testing on several samples, the typical maximum all-core frequency on a 2700X is around 4.05GHz.
The variation is 3.975GHz - 4.075GHz, depending on the specimen.

I'm confused by this line of conversation. This is with PBO disabled (so keeping the TDP as upper boundary), isn't it?

 

[Zucker2k]

Ryzen tends to ship with coolers that meet the CPUs advertised specs, and then says that a better cooling solution should provide better results, even if not much better.
This 9900k ships without a cooler but states 95W TDP. If you put a 95W cooler with it, it is never going to hit that ACT for even a second. In fact, I saw some links to reviews claiming that it wouldn't even post.
The lesson is simple; ship a cooler that allows the CPU to hit the rated specs. Right now, none of us know what cooler is the minimum required. Nor do we know how a 9900k performs with it.
Sure, no-one will ever use that cooler (or so you'll claim), but what relevance does that have when 720p low are used as a basis for CPU bound gaming tests?
The reviews so far suggest that the regular Joe is going to get punked by this 9900k. That's not good for longterm mindshare.

The 2700x lacks an IGP but you hardly hear about that? The 9900k is a chip pregnant with and igp and AVX2 and still creaming everything in sight. My point is, you don't buy a Bugatti and fill it with cheap fuel. You fill it with expensive fuel for optimal performance. The IGP is the icing on the cake.

 

[TheELF]

Guess it depends on what resolution and settings you play at. According to TPU, at 1440p a 9900k is 4% faster than a 2700X and at 4k that dwindles to 1.6%. If you game on a higher res ultrawide (3440x1440), that puts the 9900k at ~3% faster. I would venture to bet that 1.6-4% difference is indistinguishable on average. The 9900k's lead will increase with a 2080Ti (TPU used a 1080Ti) but the difference won't be huge.

But I agree, if you play at 1080p on medium settings with a 2080Ti, the 9900k is the chip for you.

Spoiler

Well if you play at 4k the 2200g is 88% and the g4560 is 92% of the 9900k ,so what does that mean?

 

[TheELF]

So in summary, the numbers are all over the place.
Is it any wonder why Intel expected to get away with the PT FUD?

The man on the street buys a 9900k, plugs it into his existing motherboard, and either it fails to boot or it gives worse performance than what he was replacing. It seems as though without exactly the right components you'll never get this CPU running at the performance level that it is being advertised as

Cough cough,so wait the PT is FUD because the man on the street knows that he can't populate all mem slots or that he has to get a special brand/model or that he can't run a manufacturers software that states game mode?
But the power issue isn't FUD because we can't expect the man on the street to be able to differentiate between a gaming and a eco mobo.

 

[IntelUser2000]

All of a sudden, XFR and Turbo Boost are bad.

No one has, ever, said that.

What we are saying is pretending 95W is the TDP while really using 150W all the time is bad. To meet 95W TDP, Turbo has to be off.

Do you really have a hard time understanding this?

Previously: 95W meant the CPU will take treat that as the sustained maximum power. The 95W and the PCU(Power Control Unit) intelligently decided what the final clocks are going to be. In very intensive scenarios it would likely be at base. Most of the time, it has enough headroom that it'll reach max Turbo.

Now: 95W now means when a crappy HSF makes it highly thermally limited preventing it from going further or in extremely light and bursty scenarios where you'd probably be just as well off using a 15W U chip.

The 2700x lacks an IGP but you hardly hear about that?

I'm not seeing him anywhere talking about the iGPU.

you can bet intel seeded golden samples to reviewers.

Remember Tomshardware's "Just buy it" article with the Nvidia RTX series? The worse thing is some reviewers at TH didn't like the RTX and it was shown in their article. Meaning within TH, there are disagreements. Must be fun to work there.

All types of press are shady nowadays, politics and non-politics. Their views and opinion pieces are entirely shaped by the companies that give them money.

What happened to reporting the truth?

 

[Zucker2k]

No one has, ever, said that.

What we are saying is pretending 95W is the TDP while really using 150W all the time is bad. To meet 95W TDP, Turbo has to be off.

Do you really have a hard time understanding this?

Previously: 95W meant the CPU will take treat that as the sustained maximum power. The 95W and the PCU(Power Control Unit) intelligently decided what the final clocks are going to be. In very intensive scenarios it would likely be at base. Most of the time, it has enough headroom that it'll reach max Turbo.

Now: 95W now means when a crappy HSF makes it highly thermally limited preventing it from going further or in extremely light and bursty scenarios where you'd probably be just as well off using a 15W U chip.



I'm not seeing him anywhere talking about the iGPU.



Remember Tomshardware's "Just buy it" article with the Nvidia RTX series?

All types of press are shady nowadays, politics and non-politics. Their views and opinion pieces are entirely shaped by the companies that give them money.

What happened to reporting the truth?

No.

Do you understand that Turbo is there to take advantage of certain preconditions?

Edit: Where is your outrage over the 2700x and PBO vis a vis the 105watts?

 

[inf64]

@IntelUser2000

What do you think about these CPUs working under even "lighter" loads but in summer 2019? There will be a lot of "sweating" going around in rooms with these little ovens. I wonder if the best of the best AIO WC will be able too cool these things down in June-August weather. Fun times ahead.

 

[IntelUser2000]

No.

Do you understand that Turbo is there to take advantage of certain preconditions?

Edit: Where is your outrage over the 2700x and PBO vis a vis the 105watts?

Do you? Turbo doesn't mean lying about TDP figures. It never was, until recently.

The companies that lead the market set the basis for everyone. The market leader is Intel, so whatever Intel does wrong likely competitors like AMD will follow. Because the smaller company has no choice. If the leader goes into the gutter the rest follows.

What do you think about these CPUs working under even "lighter" loads but in summer 2019?

Why don't you ask the 9900K as I'm not a CPU?

Ovens use 1000 watts, which is 6-7 times the power stock loaded 9900K uses. You'd also want to note most high end CPUs are in the "oven" range you are talking about.

 

[SirDinadan]

The 2700x lacks an IGP but you hardly hear about that? The 9900k is a chip pregnant with and igp and AVX2 and still creaming everything in sight. My point is, you don't buy a Bugatti and fill it with cheap fuel. You fill it with expensive fuel for optimal performance. The IGP is the icing on the cake.

How does this make any sense? Seriously!
IMO, the IGP on Intel's high-end CPUs is just a waste of die space. For gaming, this is definitely true, for compute maybe you can find some use-cases, but generally, there are very few applications that can take advantage of Intel's IGPs.

And the cheap fuel metaphor is just bad. Intel's IGP solution is inferior to AMD's regarding performance. Yes, Pinnacle Ridge doesn't have any - but is there anyone out there who uses a high-end CPU with integrated graphics?

And I don't have a problem with your supercar analogy. The 9900K is the ultimate gaming CPU.

 

[jpiniero]

Intel's IGP is more than competent outside of gaming. That being said, I do see it going for K, esp once Intel makes the move to EMIB and friends.

 

[Asterox]

As expected vs Ryzen, Intel pushed i9 9900K to the limit or "soldering for overclocking very small gains".

 

[Kenmitch]

The 2700x lacks an IGP but you hardly hear about that? The 9900k is a chip pregnant with and igp and AVX2 and still creaming everything in sight. My point is, you don't buy a Bugatti and fill it with cheap fuel. You fill it with expensive fuel for optimal performance. The IGP is the icing on the cake.

Bringing up the iGP card? Nobody buys a 9900k or 2700x class cpu to play on a iGPU. The only benefit of the iGPU in the 9900k is it'll be dead silicone for the majority of end users who'll purchase the product....Guess it's were those 720p gaming benchmarks come into play.

 

[IntelUser2000]

How does this make any sense? Seriously!
IMO, the IGP on Intel's high-end CPUs is just a waste of die space.

This is a very often said and misguided belief.

The reason their highest end consumer chip has an iGPU is because every other chip is cut down from it. From the 8+2(8 CPU cores + GT2 GPU core) they make 6+2, 4+2, 2+2, and 2+1.

Pick up a flyer from a newspaper and you can see systems featuring high end chips using only iGPUs. Many people are out there that wants and/or needs the fastest GPU without needing a discrete GPU. iGPUs are perfect for them. If they take out the iGPU then it'll be artificially locking such markets out. Even among the people that add graphics cards, the iGPU serves as a great troubleshooting tool, and multi-monitor setup. It also allows them to use the computer for the short time they are waiting for the GPU to arrive.

While cutting to make a smaller chip is easier than making a bigger one, they still need to justify the existence of such a chip. That market will be a lot smaller. If you are suggesting they make a 10+0 chip, it won't happen without having a 10+2 chip and you'll complain again. They'll not consider making an iGPU-less die because the effort and cost involved is greater than having it in.

 

[jpiniero]

They'll not consider making an iGPU-less die because the effort and cost involved is greater than having it in.

That's where EMIB comes in.

 

[IntelUser2000]

That's where EMIB comes in.

EMIB is NOT a low cost solution. It is a lower cost alternative to silicon interposers, where a lot of bandwidth is required. Intel does not use EMIB on the U-parts with the on-package chipset because the bandwidth requirements are very little.

Lower end dies like current dual cores will continue to be monolithic because monolithic dies are cheaper overall when the die sizes are small.

 

[SirDinadan]

Intel's IGP is more than competent outside of gaming.

Feature-wise they are top-notch and drivers have improved a lot, but can you share some applications where Intel's IGPs can compete with AMD's on-board gfx? Nearly 2 years ago a new driver code-branch improved the performance of the IGPs a lot in LuxMark and with that it outperformed Kaveri. I would love to test some apps with current IGPs vs AMD.

 

[Arachnotronic]

Bringing up the iGP card? Nobody buys a 9900k or 2700x class cpu to play on a iGPU. The only benefit of the iGPU in the 9900k is it'll be dead silicone for the majority of end users who'll purchase the product....Guess it's were those 720p gaming benchmarks come into play.

The 9900K is based on a die that will go into a wide range of systems like business PCs and typical consumer PCs, and those machines overwhelmingly rely on integrated graphics.

 

[IntelUser2000]

Feature-wise they are top-notch and drivers have improved a lot, but can you share some applications where Intel's IGPs can compete with AMD's on-board gfx?

Are you talking about in 2D? Because they are quite good there. In 3D AMD iGPUs are much better.

Then again, Skylake's HD 530 was fairly decent. The problem is that was back in 2015. It's 3 years later and they are using the same iGPU. For some reason it was said GPU was the most problematic part for Cannonlake. That's why we see it without the GPU.

 

[KompuKare]

That is a good question. :thumbsup:
It may not have any serious impact while gaming, but it is good to see how the fix affects theoretical performance versus 8 series , patched and unpatched.

Well, while a bit of a gaming niche, Fallout 4 with tons of mods can be quite I/O intensive as it streams stuff in. So much so that users say they've certainly noticed the patches impacting they're experience.

 

[jpiniero]

Lower end dies like current dual cores will continue to be monolithic because monolithic dies are cheaper overall when the die sizes are small.

The flexibility and scalibility will make it worth it. I don't think Intel would have dared go after the dGPU market if they didn't have a scalable GPU design ready to go with Gen12 for instance.

Feature-wise they are top-notch and drivers have improved a lot, but can you share some applications where Intel's IGPs can compete with AMD's on-board gfx? Nearly 2 years ago a new driver code-branch improved the performance of the IGPs a lot in LuxMark and with that it outperformed Kaveri. I would love to test some apps with current IGPs vs AMD.

Just standard 2D and like QuickSync (which things like Webcams should be using for encoding the video). Raw compute is OK but Raven Ridge is I'm sure decently faster if only because AMD is giving much more cores.

 

[french toast]

The 9900K is based on a die that will go into a wide range of systems like business PCs and typical consumer PCs, and those machines overwhelmingly rely on integrated graphics.

Yes but we are talking about the 9900k specifically..and another poster was using this bundled waste of silicon to justify 9900ks price and value, the igpu offers next to no value to the 9900k what so ever and shouldn't even be mentioned.
If you are buying a 9900k to play space invaders in 2k then you need your head testing.

 

[The Stilt]

I'm confused by this line of conversation. This is with PBO disabled (so keeping the TDP as upper boundary), isn't it?

The limits do not affect the maximum frequencies.
They can be lower than the maximum, but not higher.

Even if you run 2700X with LN2 and disable all of the power / current limits, it will not boost higher than e.g. 4.35GHz for the best two cores of the CPU.

4.35GHz for the best two cores of the CPU (marked with a golden and silver star in Ryzen Master), 4.2GHz for the rest (1-2C load).
4.075GHz for all cores, unless limited by PPT, TDC, EDC, thermal or reliability (FIT). Clock reductions starts at 85°C (95°C tCTL), unless configured to a lower value.

The power management must be reconfigured in order to allow higher frequencies, my "eXFR" ("Performance Enhancer" on ASUS boards) does just that.

 

[jpiniero]

Yes but we are talking about the 9900k specifically..and another poster was using this bundled waste of silicon to justify 9900ks price and value, the igpu offers next to no value to the 9900k what so ever and shouldn't even be mentioned.
If you are buying a 9900k to play space invaders in 2k then you need your head testing.

Until EMIB arrives though...

Intel looks like it is going to release 8 core "mobile" parts. So the IGP will be useful there.

 

[french toast]

The 2700x lacks an IGP but you hardly hear about that? The 9900k is a chip pregnant with and igp and AVX2 and still creaming everything in sight. My point is, you don't buy a Bugatti and fill it with cheap fuel. You fill it with expensive fuel for optimal performance. The IGP is the icing on the cake.

I am sorry but this whole post is complete nonsense...when you are bigging up a ~$600 processor about being 'pregnant with an igpu' that it is the 'icing on the cake' and 'creams everything' ...needing 'expensive fuel'...
Honestly...

 

[William Gaatjes]

Well, while a bit of a gaming niche, Fallout 4 with tons of mods can be quite I/O intensive as it streams stuff in. So much so that users say they've certainly noticed the patches impacting they're experience.

Interesting. Never thought of that.