16" MBP Geekbench scores so far show big differences

[wallysb01]

So it seems we have the BIG question: does the 16 vs 32GB affect the big differences between i9 scores

I don’t think that’s possible with what geekbench is doing unless Apple did something really weird and only populated one DIMM slot in the 16GB version, which would crimp the potential data flow to the CPU. Geekbench’s CPU test should be otherwise unimpaired by the amount of RAM available.

 

[The Mercurian]

In all seriousness - why?

I've found the past few generations to be more then up to snuff performance wise - of course, my needs are not the same as your.

Eh fair question. Basically for faster chips. I guess I'm just assuming that with a die shrink they will be a chunk faster. Aloso I was thinking it would be good to wait a few months and see if any problems crop up on these new machines. But this is perhaps wishful thinking i'm very tempted not to wait 😅

 

[mmjrk]

What’s bothering me a bit is that only 2.3 GHz models do not seem to benefit improved thermal management compared to 15 inch models. Quite the opposite actually, scores seem lower. Makes me wonder whether Apple limits its performance on purpose to better differentiate 2.4 GHz models. Makes me regret I went for 9880H...

 

[The Mercurian]

I don’t think that’s possible with what geekbench is doing unless Apple did something really weird and only populated one DIMM slot in the 16GB version, which would crimp the potential data flow to the CPU. Gerkbench’s CPU text should be otherwise unimpaired by the amount of RAM available. EDIT: oops no die shrink. But I guess if I splurge now and in 6 months there is a faster machine I'll be annoyed

Fair point. I guess I'm trying to understand what is the explanation for the fairly wide spread of scores on the machines? I guess maybe some people are runnign them while doing other things 🤷‍♂️
[automerge]1573923606[/automerge]

What’s bothering me a bit is that only 2.3 GHz models do not seem to benefit improved thermal management compared to 15 inch models. Quite the opposite actually, scores seem lower. Makes me wonder whether Apple limits its performance on purpose to better differentiate 2.4 GHz models. Makes me regret I went for 9880H...

But the slower chips didn't have such thermal problems in the first place. The improved thermal managment should benefit the hotter chips more.

 

[wallysb01]

In all seriousness - why?

I've found the past few generations to be more then up to snuff performance wise - of course, my needs are not the same as your.

And next gen H series chips are still going to be on 14nm. Things like LPDDR4X or Wifi6 will be available, but CPU wise it’s likely to be a very marginal upgrade. I don’t think other upgrades that ice lake are getting, like AVX-512 will be available, for example.

Now seems like a good time to buy to me, but if you can wait, waiting will always bring a better machine. The big upgrade though will be when 10nm finally makes it way to H series chips, but that still 2 years away. I can’t wait that long. So I’ll take what I can get now. Knowing I’m missing more efficient and faster RAM, but oh well, have to pull the trigger eventually and no matter when you do, something better is coming soon.

 

[Mojo1019]

Well damn. I have buyer’s remorse now.

I am sure you can refund it! This just means you will have to wait more to have your new baby.

 

[The Mercurian]

And next gen H series chips are still going to be on 14nm. Things like LPDDR4X or Wifi6 will be available, but CPU wise it’s likely to be a very marginal upgrade. I don’t think other upgrades that ice lake are getting, like AVX-512 will be available, for example.

Now seems like a good time to buy to me, but if you can wait, waiting will always bring a better machine. The big upgrade though will be when 10nm finally makes it way to H series chips, but that still 2 years away. I can’t wait that long. So I’ll take what I can get now. Knowing I’m missing more efficient and faster RAM, but oh well, have to pull the trigger eventually and no matter when you do, something better is coming soon.

Is there any chance of core count increase? Will we see ice lake laptop chips with 10 cores for example ?

 

[intrepidcase]

But the slower chips didn't have such thermal problems in the first place. The improved thermal managment should benefit the hotter chips more.

On your first point I would disagree. Check out the new Maxtech review of the i7 2.6 16" vs i7 2.6 15" ... the improved thermals clearly improved performance with the same baseline chip.

But I would agree on the second point - there should be even more benefit for the faster chips ... which is maybe why the i9 2.4 seems to be benchmarking a bit better than nominal clock speed increase?

[automerge]1573923907[/automerge]

 

[wallysb01]

Is there any chance of core count increase? Will we see ice lake laptop chips with 10 cores for example ?

we will see Comet lake cores (14nm 10th gen), which sounds like will bring the possibility of 10 cores in a laptop. I’m curious how the power management of that goes though. It’s obviously worth it for 6 -> 8 cores as we are seeing in this generation, but without a node shrink, the added value of more cores will probably diminish to some extent. Maybe other platform changes will help. Anyway, it will be some improvement, but more of a normal 5-10%, rather than a larger step like what the move to 10nm would bring (particularly tasks supporting AVX-512).

 

[iMi]

I second the idea that waiting for next year’s chips doesn’t make sense. Of course they will be faster and then slower than the chips that will supersede it. The reality is that any of the past two or three generations are pretty powerful and should be sufficient for a vast majority of professional applications.

At some point, you’re just chasing incremental improvements over the loss of great performance today. Plus, as it has been said so many times before on this forum, buy what you need now. If you need it. Playing the “next year’s will be better” is a never ending cycle of anticipation, purchase, regret.

 

[gplusplus]

I am sure you can refund it! This just means you will have to wait more to have your new baby.

Well I was lucky enough to cancel it while it was still in the processing stage. Sweet! I don’t mind waiting another couple weeks, considering I plan on going at least 4 years on this thing.

I ordered maxed out everything other than storage; I’m at 2 TB on the storage.

What I got is definitely overkill for most people, but as many reviewers have said, “you’ll already know if you need 64 gb of memory. Otherwise, you don’t.” I already come close to maxing out 32gb at times with some workloads, so I think it’s safe to say that I’ll need more within a couple years or so. Can’t upgrade later and $400 isn’t too significant, considering that my configuration with AppleCare and taxes exceeded $5k.
[automerge]1573927471[/automerge]

Why which one did you buy?

I originally ordered the 2.3 ghz i9, but I just canceled and re-ordered the 2.4 ghz. That officially puts me at an “everything maxed out except for storage” configuration. I’m at 2 TB on the storage front.

 

[Whackman]

So... hmn, i ordered the 2.3GHz 8 core version with 32GB RAM.
Is that a bad choice now? Because people said i would never notice any difference between the 2.3GHz and the 2.4Ghz..

 

[wallysb01]

Well I was lucky enough to cancel it while it was still in the processing stage. Sweet! I don’t mind waiting another couple weeks, considering I plan on going at least 4 years on this thing.

I ordered maxed out everything other than storage; I’m at 2 TB on the storage.

What I got is definitely overkill for most people, but as many reviewers have said, “you’ll already know if you need 64 gb of memory. Otherwise, you don’t.” I already come close to maxing out 32gb at times with some workloads, so I think it’s safe to say that I’ll need more within a couple years or so. Can’t upgrade later and $400 isn’t too significant, considering that my configuration with AppleCare and taxes exceeded $5k.
[automerge]1573927471[/automerge]

I originally ordered the 2.3 ghz i9, but I just canceled and re-ordered the 2.4 ghz. That officially puts me at an “everything maxed out except for storage” configuration. I’m at 2 TB on the storage front.

good choice I think. And IMO anything above 2TB is a bit much. (I went with the 1TB, but 2 did have me tempted.) You pay a lot for that marginal 2 or 6 TBs and for most people it probably only serves to enable poor decision making on data storage solutions. Hoarding a lot of data on a laptop just seems risky to me and unless you for some reason have absolutely insane working file sizes completely unnecessary as well.
[automerge]1573928777[/automerge]

So... hmn, i ordered the 2.3GHz 8 core version with 32GB RAM.
Is that a bad choice now? Because people said i would never notice any difference between the 2.3GHz and the 2.4Ghz..

nothing is necessarily a bad choice, but if you can really put that 8 core 2.3 to work, then you’ll probably be benefited by the upgrade to 2.4 as well. And for $200, its not a huge cost to incur.

 

[Whackman]

good choice I think. And IMO anything above 2TB is a bit much. (I went with the 1TB, but 2 did have me tempted.) You pay a lot for that marginal 2 or 6 TBs and for most people it probably only serves to enable poor decision making on data storage solutions. Hoarding a lot of data on a laptop just seems risky to me and unless you for some reason have absolutely insane working file sizes completely unnecessary as well.
[automerge]1573928777[/automerge]


nothing is necessarily a bad choice, but if you can really put that 8 core 2.3 to work, then you’ll probably be benefited by the upgrade to 2.4 as well. And for $200, its not a huge cost to incur.

Not sure. I'll try it when i get it but i imagine the difference coming from a quadcore 2.6 GHz i7 and 16 GB RAM will be pretty big anyway (i have a mid 2012 MacbookPro now).

 

[athanis]

So... hmn, i ordered the 2.3GHz 8 core version with 32GB RAM.
Is that a bad choice now? Because people said i would never notice any difference between the 2.3GHz and the 2.4Ghz..

I don't think it's a bad choice.


Updated median:

	samples	median single-core	median multi-core
i7	69​	1059​	5589​
i9 2.3	72​	1067​	6497​
i9 2.4	32​	1125​	7235​

So, as you can see the i9 2.3 is now just ~5 % slower in single-core performance (compared to ~9 % slower in my previous post with less samples) and ~11 % slower in multi-core performance (again, compared to ~13 % in my previous post).

I think, the more samples we get the smaller the distance between the i9 2.3 and i9 2.4 becomes, which leads to a more realistic overall picture.

 

[danwells]

Given that the 2.4 gHz i9-9880HK is almost certainly just a binned version of the 2.3 gHz i9-9880H (I suspect that the overclock capability of the HK can be turned on or off after testing at Intel), could the binned processor be handling undervolting better (need less voltage to hit a given speed)?

If Intel tests each chip to see if it meets specs as an i9-9880HK first, that means that the 2.3 gHz chips are all "losers" in the silicon lottery. It's also possible that some of the 2.3 gHz chips don't meet the 2.4 gHz specs while others weren't tested at the higher speeds due to relative demand...

 

[JustHereforHelp69]

Well I was lucky enough to cancel it while it was still in the processing stage. Sweet! I don’t mind waiting another couple weeks, considering I plan on going at least 4 years on this thing.

I ordered maxed out everything other than storage; I’m at 2 TB on the storage.

What I got is definitely overkill for most people, but as many reviewers have said, “you’ll already know if you need 64 gb of memory. Otherwise, you don’t.” I already come close to maxing out 32gb at times with some workloads, so I think it’s safe to say that I’ll need more within a couple years or so. Can’t upgrade later and $400 isn’t too significant, considering that my configuration with AppleCare and taxes exceeded $5k.
[automerge]1573927471[/automerge]

I originally ordered the 2.3 ghz i9, but I just canceled and re-ordered the 2.4 ghz. That officially puts me at an “everything maxed out except for storage” configuration. I’m at 2 TB on the storage front.

I did exactly what you did, just canceled mine and got the 2.4, I am going to have to wait like another week or two but I can live with that. Hopefully some third party cases come out by the time it comes because that is something I am worried about when traveling with it.

 

[Spectrum]

Yeah, I am wondering about this but one would assume it didn't? All/most of the i9 2.3 scores have 16GB of ram, and all/most of the i9 2.4 scores have 32GB of RAM. Regardless, I am getting 32GB no matter what.

Increasing RAM form 8 GB to 32GB increased Geekbench scores for my 2018 mac mini...
Not sure why...but others have reported it too.

 

[Fingerprintt]

I don't think it's a bad choice.


Updated median:

	samples	median single-core	median multi-core
i7	69​	1059​	5589​
i9 2.3	72​	1067​	6497​
i9 2.4	32​	1125​	7235​

So, as you can see the i9 2.3 is now just ~5 % slower in single-core performance (compared to ~9 % slower in my previous post with less samples) and ~11 % slower in multi-core performance (again, compared to ~13 % in my previous post).

I think, the more samples we get the smaller the distance between the i9 2.3 and i9 2.4 becomes, which leads to a more realistic overall picture.

Thanks so much for the update, seems we're now really back to those Numbers after all...

 

[gplusplus]

I did exactly what you did, just canceled mine and got the 2.4, I am going to have to wait like another week or two but I can live with that. Hopefully some third party cases come out by the time it comes because that is something I am worried about when traveling with it.

I’m with you on that. Although I think I’d rather go for a clear skin, but still, it gives us some time to wait for products in general.

 

[Macshroomer]

good choice I think. And IMO anything above 2TB is a bit much. (I went with the 1TB, but 2 did have me tempted.) You pay a lot for that marginal 2 or 6 TBs and for most people it probably only serves to enable poor decision making on data storage solutions. Hoarding a lot of data on a laptop just seems risky to me and unless you for some reason have absolutely insane working file sizes completely unnecessary as well.

LOL, X1000

Why on EARTH do people have to constantly justify their purchase by questioning what someone else might choose??

I run a very high end photography business and within that scope I occasionally direct and produce video content. That means I don't do the heavy lifting of motion camera work but subcontract it out and then ingest all the files. A single day shoot on an ad campaign can net 150-300GB of files. Multiply that by 2-3 with an out of town location shoot and we are easily at 1-1.8TB of files on a shoot. Then process those files on location to make the A.D. extra happy and I might have a stitched file for an airport boarding area display that needs to be 30' feet at 200DPI...that is like a 1.5-2GB file, for a single photo.

So what I ordered is the 2.4 / 64GB / 8GBVRAM / 4TB machine. I will use that 4TB with a pair of 2TB OWC Envoy PCIE externals and be just fine thank you every much. I always keep at least 25% headroom on any drive so that leaves me with 3TB of usable.

If 8TB were as cheap an upgrade as 4 was, I would not have thought twice about doing it.

 

[wallysb01]

LOL, X1000

Why on EARTH do people have to constantly justify their purchase by questioning what someone else might choose??

I run a very high end photography business and within that scope I occasionally direct and produce video content. That means I don't do the heavy lifting of motion camera work but subcontract it out and then ingest all the files. A single day shoot on an ad campaign can net 150-300GB of files. Multiply that by 2-3 with an out of town location shoot and we are easily at 1-1.8TB of files on a shoot. Then process those files on location to make the A.D. extra happy and I might have a stitched file for an airport boarding area display that needs to be 30' feet at 200DPI...that is like a 1.5-2GB file, for a single photo.

So what I ordered is the 2.4 / 64GB / 8GBVRAM / 4TB machine. I will use that 4TB with a pair of 2TB OWC Envoy PCIE externals and be just fine thank you every much. I always keep at least 25% headroom on any drive so that leaves me with 3TB of usable.

If 8TB were as cheap an upgrade as 4 was, I would not have thought twice about doing it.

Um, dude. You just described the exact kind of situation I used as an exception.... “unless you for some reason have absolutely insane working file sizes”...

like I have said before here. No configuration is absolutely a bad choice, only bad fits for a given workflow. If you actually need that space great, get it. And it’s wonderful that Apple is giving you this option.

glad to help you get LOL though...

 

[leman]

While sipping my morning tee I had some inspiration, so I have parsed all the currently available Geekbench 5 results for the 16" MBP — see the results below. Some explanation: the dots represent the individual score, the lines are the 25%, 50% (median) and 75% quantiles (that is, everything between the top and bottom line is where 50% of the relevant scores are), and the shapes represent the density distribution of the scores (so called violin plot, which are my favourites)

In addition, I ran a simple unequal variance T-test on the data, with the following result:

- assuming 5% significance threshold, there is no statistical difference between the single-core results for i9-9880H and i9-9980HK, albeit the i9-9980HK is a bit faster on average (95% confidence interval of score differences between -5.52462 and 82.82824 )

- assuming 5% significance threshold, there is a statistical difference between the multi-core results for i9-9880H and i9-9980HK (95% confidence interval of score differences between 497.2288 and 746.6590)

Bottomline: i9-9980HK is around 10% faster on average for multi-core (and 3% faster on average for single-core). This is quite unexpected given the CPU specs. The HK CPUs are rigorously binned and/or Apple might be undervolting them. At any rate,I am starting to regret that I ordered the i9-9880H, it might need to go back 😅

EDIT: added RAM, as requested by @The Mercurian . I also did a quick regression analysis on this, and RAM does not make any difference in a regression model. But at the same time, we can't really analyse the impact of RAM as we don't the enough data points: users who buy the high-end CPU also never seem to get 16GB RAM. Here is a table:

CPU RAM n
i7-9750H 16384 MB 109
i9-9880H 16384 MB 87
i9-9980HK 32768 MB 40
i9-9980HK 65536 MB 1

 

[The Mercurian]

Nice work @leman - I love the geom_violin() 😅 Did you pull the data off with a script or manually? If script can you modify it to also pull the RAM in the machine? I'm very curious if any of the spread is due to whatever RAM is in the machines!

 

[leman]

Nice work @leman - I love the geom_violin() 😅 Did you pull the data off with a script or manually? If script can you modify it to also pull the RAM in the machine? I'm very curious if any of the spread is due to whatever RAM is in the machines!

I pulled the data from Geekbench browser and wrote a crappy HTML parser to extract the tables, but unfortunately, there is no RAM info on those pages. One could follow the links to the individual machines and so some more parsing but unfortunately my tea time is over