The upgraded Core i9 (2.4 GHz) for the 16" MBP is kind of a bad deal

[sebpettersson]

The base i9 model uses Core i9-9880H and the upgraded one uses Intel Core i9-9980HK (someone correct me if I'm wrong, but those are the ones with specs that match).

Intel's listed recommended customer price is $556.00 for the first part and $583.00 for the second. That $27 difference turn into a $200 dollar difference when upgrading the 16" MBP on the Apple store. Now keep in mind that Apple most likely pays nowhere near those prices for the volumes they are ordering, and in reality the difference should be less than $27.

The upgrade cost is almost pure profit margin for Apple.

Edit: I update the thread title to make it more clear I talking about the higher tier i9 and not the i7 base model vs the i9 model.

 

[maflynn]

Are you taking into account that the 8 core MBP also comes with a 1TB drive instead of 512GB, or that the GPU is a 5500M vs. a 5300M?

 

[sebpettersson]

I'm talking about the difference between the two i9 models, not the i7 vs i9.

 

[maflynn]

I'm talking about the difference between the two i9 models, not the i7 vs i9.

If you don't think its a good deal, then don't get it ¯\_(ツ)_/¯

I usually only buy the base model myself because those CPU (and even GPU) upgrades will be under utilized by me.

 

[cappo3]

I'm talking about the difference between the two i9 models, not the i7 vs i9.

View attachment 877036

I've seen this and thought the same myself. Is there any other difference between the two CPUs, apart from the slight frequency bump?

 

[leman]

Intel's listed recommended customer price is $556.00 for the first part and $583.00 for the second. That $27 difference turn into a $200 dollar difference when upgrading the 16" MBP on the Apple store.

The recommended price is just a number, not to mention that this is volume pricing (Intel states that RCP typically applies to batches of 1000 or more). We don't know how much Apple is really paying for them and how many of these CPUs even are out there (they seem to be fairly rare). I wouldn't interpret that much into Intel RCP, simply because there is no way for you to get this CPU separately.

Now, I agree that the i9-9980HK is a bad deal, but for a different reason. There is just not enough performance difference to justify the extra $200. The base clock is meaningless — both will perform better than that and both will plateau at approx. the same frequency, since they are thermally limited by the chassis. And you can forget about seeing thermal velocity boost in practice, so those 5.0Ghz won't ever be reachable either.

 

[theluggage]

I'm talking about the difference between the two i9 models, not the i7 vs i9.

Well, yes, considering the difference between Intel's "recommended customer price" for the two chips is actually about $30 - but if you were expecting Apple to offer the upgrade for, say, a mere 50% markup (or, as you or I would probably do if building a $2800 computer, just stick in the more powerful chip and offset the $30 difference against the logistical cost of each upgrade option) then you must be new here

If you're buying a computer to use then I wouldn't worry about it & spend the $200 on RAM or storage. If you want it to win a game of Top Trumps (which seems to be the current MR definition of "pro") then just pay the tax...

 

[CHA05 R31GN5]

The recommended price is just a number, not to mention that this is volume pricing (Intel states that RCP typically applies to batches of 1000 or more). We don't know how much Apple is really paying for them and how many of these CPUs even are out there (they seem to be fairly rare). I wouldn't interpret that much into Intel RCP, simply because there is no way for you to get this CPU separately.

Now, I agree that the i9-9980HK is a bad deal, but for a different reason. There is just not enough performance difference to justify the extra $200. The base clock is meaningless — both will perform better than that and both will plateau at approx. the same frequency, since they are thermally limited by the chassis. And you can forget about seeing thermal velocity boost in practice, so those 5.0Ghz won't ever be reachable either.

This. Any benchmarks you find of these CPUs will probably not take into account the tighter thermal limits of MBPs over other solutions implemented by manufacturers. Totally agree, that a $200 upgrade isn't justified for what's gained between the two.

 

[leman]

This. Any benchmarks you find of these CPUs will probably not take into account the tighter thermal limits of MBPs over other solutions implemented by manufacturers. Totally agree, that a $200 upgrade isn't justified for what's gained between the two.

To add to this, the difference between the two i9 models will only start becoming clear in laptops that offer 100W of cooling capacity (since you'd need to essentially run them as desktop CPUs) — and there are not that many. Appel claims to have improved cooling capacity of the MBP by 12Watt — which is great, but it will still only put it at 75-80W sustained thermal output.

 

[Sirmausalot]

It's only $150 upgrade with educational pricing. Since I do a LOT of encoding, I decided to take the chance. I hope the improved cooling allows it to reach the higher speeds more consistently.

 

[wallysb01]

Everyone needs to be aware that the 2.3 vs 2.4 difference is there because the individual chips on the 2.4 passed validation at a higher level. Yes, the MBP will be thermally limited, but intel has more aggressively clocked them and the 2.4 will have better performance in the same power envelope. We saw this with the mid-2019 machine as well, which was even more thermally limited.

Not saying the $200 upgrade is or isn't worth it to you, but there will be a difference in performance. It will be small, but it will be there.

Also remember for tasks that don't run all 8 cores, you likely won't be thermally limited, but the 2.4 will clock all the way up to 5GHz, while the 2.3 will get you up to 4.8. At what core utilization levels that stops being true will have to wait and see.

Also, here's a bit of a comparison of the mid-2019 15" 2.4 vs the late-2019 16" 2.3: https://www.macworld.co.uk/review/mac-laptops/macbook-pro-16in-3777729/

The 16" 2.3 is besting the 2.4 in the 15". So its starting to look like this new thermal solution is a noticeable improvement. It will be nice to see more thorough comparisons.

 

[leman]

Everyone needs to be aware that the 2.3 vs 2.4 difference is there because the individual chips on the 2.4 passed validation at a higher level. Yes, the MBP will be thermally limited, but intel has more aggressively clocked them and the 2.4 will have better performance in the same power envelope. We saw this with the mid-2019 machine as well, which was even more thermally limited.

True, but then you have this...

That embarrassing realization when the Core i9-9880H has outperformed every Core i9-9980HK laptop thus far

The slowest laptop with the Core i9-9980HK is the Dell XPS 15 7590 which the Core i9-9880H can outperform by as much 25 percent. If you're out looking for the "ultimate" Core i9 performance on mobile, the i9-9980HK offers only a marginal boost over the fastest i9-9880H laptop in the best of...

www.notebookcheck.net

[automerge]1573748824[/automerge]

Also, here's a bit of a comparison of the mid-2019 15" 2.4 vs the late-2019 16" 2.3: https://www.macworld.co.uk/review/mac-laptops/macbook-pro-16in-3777729/

The 16" 2.3 is besting the 2.4 in the 15". So its starting to look like this new thermal solution is a noticeable improvement. It will be nice to see more thorough comparisons.

From this, it looks that the CPU performance did not change much. Larger gaming laptops still get around 15-20% more performance out of the same CPU. But of course we need to do more tests, plus, I think there might be an issue with R20 under macOS.

 

[wallysb01]

True, but then you have this...

That embarrassing realization when the Core i9-9880H has outperformed every Core i9-9980HK laptop thus far

The slowest laptop with the Core i9-9980HK is the Dell XPS 15 7590 which the Core i9-9880H can outperform by as much 25 percent. If you're out looking for the "ultimate" Core i9 performance on mobile, the i9-9980HK offers only a marginal boost over the fastest i9-9880H laptop in the best of...

www.notebookcheck.net

[automerge]1573748824[/automerge]


From this, it looks that the CPU performance did not change much. Larger gaming laptops still get around 15-20% more performance out of the same CPU. But of course we need to do more tests, plus, I think there might be an issue with R20 under macOS.

I'm not sure what to make of that notebookcheck article. Their own run down of the i9-9880H vs i9-9980HK shows small gains, but a fairly consistent 4-5% across multiple benchmarks. It seems that what ever is going on in the MSI machine that is used as the bench mark for the 9880H is probably the deal, not the processor itself.

It also doesn't match what we saw in the 2019 15" machine, which was small gains with the upgrade from the 2.3 to the 2.4.

As for the gains going from the 15" to the 16", I certainly wouldn't expect the new 16" MBP to compete with the larger gaming laptops. If you really need that 10-20% and don't care about macOS, have fun on those 6lb, 1+ inch thick beasts.... not to mention the rainbow colored keyboards.

 

[Thysanoptera]

I'm not sure what to make of that notebookcheck article. Their own run down of the i9-9880H vs i9-9980HK shows small gains, but a fairly consistent 4-5% across multiple benchmarks. It seems that what ever is going on in the MSI machine that is used as the bench mark for the 9880H is probably the deal, not the processor itself.

It also doesn't match what we saw in the 2019 15" machine, which was small gains with the upgrade from the 2.3 to the 2.4.

As for the gains going from the 15" to the 16", I certainly wouldn't expect the new 16" MBP to compete with the larger gaming laptops. If you really need that 10-20% and don't care about macOS, have fun on those 6lb, 1+ inch thick beasts.... not to mention the rainbow colored keyboards.

For some time already notebookcheck stopped being reliable source for benchmarks. Nowadays pretty much all laptops have some user accessible power configuration options and you never know how that was set on a particular model they tested. I mean - I'm getting 20~30% different scores than they show just by flipping a switch.

Regarding 2.3 vs 2.4 on Macbook - under sustained full load they should be identical, regardless of binning (limited by thermal limits in case of apple). If they're different that means power delivery is different, and looking at various threads on this forum - they are different. 2.4 in current 15 inch is undervolted for sure, I don't know about 2.3, but it sure looks like it is not. In short - if it was a Windows machine I'd go for 2.3 no questions asked. On a Mac - unfortunately I'd be willing to pay $200 for essentially factory undervolt.

And Windows machines don't have to be larger/heavier than Macbooks - my Razer Blade can dissipate 125W between CPU+GPU with some room to spare, while my 2018 MBP maxes out at maybe 55W.

 

[Sirmausalot]

In THEORY do we have 8 cores running at 100MHz faster (800MHz) and potentially 16 threads running something like Compressor (so 1.6GHz)? Which is a potential increase of over 50% in CPU power? Though compressor also takes advantage of the T2 chip and GPU.

I'd love someone who knows their chip stuff to weigh in. I suspect around a 10% performance boost in specific applications is possible.

 

[leman]

In THEORY do we have 8 cores running at 100MHz faster (800MHz) and potentially 16 threads running something like Compressor (so 1.6GHz)? Which is a potential increase of over 50% in CPU power? Though compressor also takes advantage of the T2 chip and GPU.

That is not really how this works. The 2.3 ghz is just a lower bounds promise. Both CPU models would sustain higher clocks than that, limited by the thermal performance if the chassis. Any difference will be due to fluctuations between individual CPUs, that is, luck

I'd love someone who knows their chip stuff to weigh in. I suspect around a 10% performance boost in specific applications is possible.

If you are really lucky, you might see 5% difference either way.

 

[intrepidcase]

what to make of early geekbench scores for the two cpu's?

macbookpro16 i9-9880h - Geekbench 5 CPU Search - Geekbench

vs

macbookpro16 i9-9980hk - Geekbench 5 CPU Search - Geekbench

showing a surprising 10% difference between the two ...

 

[wallysb01]

what to make of early geekbench scores for the two cpu's?

macbookpro16 i9-9880h - Geekbench 5 CPU Search - Geekbench

vs

macbookpro16 i9-9980hk - Geekbench 5 CPU Search - Geekbench

showing a surprising 10% difference between the two ...

Geekbench is pretty "bursty". So the thermal issues might be controlled some. But Apple could be doing some additional tweaks to the 9980hk, many speculated they undervolted it in the 2019 15" version. They might be adding some type of performance adjustment here.

Also interesting that none of the i7 models are showing up yet. I guess because these are all media giveaways still and they just didn't give away the bottom model?

 

[nickf1]

This. And remember, this upgrade is 5% on top of the base price. Getting 5% more performance for 5% more money isn’t bad in my book.

 

[alphaod]

Of course going 200MHz for $200 is a bad deal. It's been like that ages. Even with desktop processors.

 

[hajime]

i9 2.4 is noisier than i9 2.3 with no better, sometime even worse, performance in most cases.

 

[Howard2k]

Cost and price are not the same.

 

[hajime]

i9 2.4 is noisier than i9 2.3 with no better, sometime even worse, performance in most cases.

Can anybody explain why? CPU lottery?

 

[Camarillo Brillo]

My 2.4 is not noisy at all.

I feel it was worth it as I’m scoring up to over 1300 single core on Geekbench, and looking at the charts I’m even beating the new Mac pros in some cases. Single core performance is important for audio production so I’m considering that an important benchmark for me. The computer does not get noisy unless the cpu is at full load, and even then it is not that loud to me, my 2016 and 2011 MBPs are both louder at full throttle.

I have read about some people having problems with excessive fan noise so I do believe that’s a problem but I don’t believe it’s the 2.4 cpu that is the problem. In fact a lot of the complaints have been from i7 owners.

 

[impulse462]

Yeah my 2.4 isn’t noisy for most stuff. Kinda surprising actually