Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
seems steep to bump up the graphics..

is there much difference between
560x vs. vega 10 vs. vega 20?

jus' askin..


50-90% increase from 560x to Vega 20.
5-15% increase from Vega 16 to 20

I would go for Vega 16. A 50% increase in price (Denmark) for the 20 is not worth for me. If I read the specs right, Vega 20 should be at most 25% better. Since you cannot upgrade, if you hope to work on this for five years, I believe is a good "investment"
 
Last edited:
My results on a 2018 i9, 32GB, 1TB, Vega 20.
Doesn't seem like a huge improvement. I had the 2016 model and was getting 13000 multicore score before. It's almost 13000 score difference between the 4 and 6 cores. Between the i9 6 cores and i9 8 cores is about 3000/4000.
 

Attachments

  • Schermata 2019-05-25 alle 12.13.53.png
    Schermata 2019-05-25 alle 12.13.53.png
    230.3 KB · Views: 216
DAMN APPLE FOR MAKING THINGS BETTER!!!!!!!

Apple waits for me to buy something before they announce upgrades.
Alienates loyal customers. They’re not managing step-ups well anymore. In 6 months they stepped-up graphics and then processors but charge the same amount. Makes it very difficult to plan, particularly for businesses.

How do you not see that?
 
My results on a 2018 i9, 32GB, 1TB, Vega 20.
Doesn't seem like a huge improvement. I had the 2016 model and was getting 13000 multicore score before. It's almost 13000 score difference between the 4 and 6 cores. Between the i9 6 cores and i9 8 cores is about 3000/4000.

Yeah. The difference between 4-core and 6-core CPU's was huge. The difference between 6-core and 8-core are small. It's probably because the 8-core CPU's are running at far lower clock speeds.

i9 (2.4 8 core) vs i9 (2.9 6 core) is basically the same performance on single core, and between 3000-4000 higher performance on multicore. When I upgraded from 4 core (2.9ghz) to 6 core (i9) I got an improvement of about 1500 on single core and 11000 on multicore. That was a big difference. 6 core to 8 core is a extremely small upgrade by comparison, and not worth it at all.
 
Eh, s/he was asking if the $200 was worth it and if they should spend it on something else? I wouldn’t rule out both or multiple upgrades based on that. But hey, feel free to walk yourself back. There were other aspects that you brought up that you are now ignoring. So hey, have a good one.

Not walking anything back...but I got lost in the weeds with our conversation, when at the end of the day, I really needed to go back to the OP’s question and make sure I had answered it to my own satisfaction. I did.

I am all for CPU upgrades where it makes sense, but this is not one of them. I believe $200 can be allocated to something with a more substantive return based on the user’s stated requirements.

My opinion is that a 4-7% increase in processing speed is eclipsed by the increases in GPU speeds in the Vega 16/20 upgrades, which vary from 35%-80% (sometimes more) based on the user’s stated workflow, other user’s who have shared similar results, the problems the Core i9-8950HK in the 2018 MacBook Pro has with throttling due to the heat generated versus the Core i9-8850H OR instead the benefit of additional DRAM would have for video and Photoshop workflows OR how additional storage might be a better investment if the user is more mobile. The OP didn’t give a lot of background, so I inferred a bit of what might benefit them more and CPU was at the bottom of that list.

Your implication that I am not thinking about this the right way or that I don’t understand what you are talking about is incredibly condescending. I understand Marginal Opportunity Cost and even if I took the time to analyze my purchasing decision using it, my gut tells me that I can take that $200 CPU upgrade cost and use it for something with a bigger bang for my buck. Not everyone is thinking about this from a work investment perspective either and budget is a huge deal for most of us...I can get a faster, more capable computer in the 2019 iMac for about the same or less money depending on upgrades I add to it. The opportunity I give up is the ability to be mobile should my clients ask me to be or if I simply want to be. At the end of the day, MacBook Pros and laptops in general are pretty bad as a primary workstation for a whole host of things and cost more for less horsepower, relative to their desktop counterparts.

Marginal Opportunity Costs sounds like a fancy way of trying to replace common sense and experience. Perhaps it gives you tangible numbers to make a dispassionate decision, but I find these decisions are rarely dispassionate, oddly enough, given that a computer is simply a tool, much the same as a hammer or a drill. Just my 2¢.
 
Not walking anything back...but I got lost in the weeds with our conversation, when at the end of the day, I really needed to go back to the OP’s question and make sure I had answered it to my own satisfaction. I did.

I am all for CPU upgrades where it makes sense, but this is not one of them. I believe $200 can be allocated to something with a more substantive return based on the user’s stated requirements.

My opinion is that a 4-7% increase in processing speed is eclipsed by the increases in GPU speeds in the Vega 16/20 upgrades, which vary from 35%-80% (sometimes more) based on the user’s stated workflow, other user’s who have shared similar results, the problems the Core i9-8950HK in the 2018 MacBook Pro has with throttling due to the heat generated versus the Core i9-8850H OR instead the benefit of additional DRAM would have for video and Photoshop workflows OR how additional storage might be a better investment if the user is more mobile. The OP didn’t give a lot of background, so I inferred a bit of what might benefit them more and CPU was at the bottom of that list.

Your implication that I am not thinking about this the right way or that I don’t understand what you are talking about is incredibly condescending. I understand Marginal Opportunity Cost and even if I took the time to analyze my purchasing decision using it, my gut tells me that I can take that $200 CPU upgrade cost and use it for something with a bigger bang for my buck. Not everyone is thinking about this from a work investment perspective either and budget is a huge deal for most of us...I can get a faster, more capable computer in the 2019 iMac for about the same or less money depending on upgrades I add to it. The opportunity I give up is the ability to be mobile should my clients ask me to be or if I simply want to be. At the end of the day, MacBook Pros and laptops in general are pretty bad as a primary workstation for a whole host of things and cost more for less horsepower, relative to their desktop counterparts.

Marginal Opportunity Costs sounds like a fancy way of trying to replace common sense and experience. Perhaps it gives you tangible numbers to make a dispassionate decision, but I find these decisions are rarely dispassionate, oddly enough, given that a computer is simply a tool, much the same as a hammer or a drill. Just my 2¢.

If you think marginal opportunity cost is just some fancy word soup that should bend over to you effector amounts to you gut feeling, I think we’re done here. The OP might very well be better served putting the first $200 towards GPU upgrades, maybe even the second. But is somehow the last $200 he’ll spend on his business cost going to be hit before he finishes configuring a pretty important tool for that business? Probably not. Thus the marginal choice.
 
I'm hoping for that, too. Geekbench or Cinebench should run a set of similar tests 3-10 times consecutively (similar to avoid caching benefits of repeating the same test) and only record the final result. A friend of mine used to work in benchmarking during the Athlon/P3 and Quake/Unreal days. He showed exactly how benchmarks cheat and why certain sites can always favor one platform over another (depending on who sponsors them). My MBP crawls after load for 10 minutes. I got a laptop cooler and it helps reduce temperature, but not enough. They're too thin and airflow is too poor.
This is why I recently got a 2019 5K iMac at home and an iMac Pro at work. I only use my 2015 MBP for emergencies when traveling or for just taking to meetings to show off mockups or whatever. They are both extremely fast and keep pace with each other, with the main difference being the iMac Pro runs quieter and has a faster GPU. I'm pretty much done with the MacBook Pro. My next portable Mac will probably be a MacBook or MacBook Air after they switch to Apple chips to use as a portable at home. Seeing how fast the iPad Pro is I'm hopeful they could make a fairly speedy MacBook to use for more casual stuff at home.
[doublepost=1558810487][/doublepost]
What workload are you using that the computer is under 20 min heavy load. Then, you probably should not be using a "laptop".
A lot of people use them to edit video. But yes, if you look in my signature I have recently moved away from using MBP for most tasks, especially since I'm doing more video work than I have ever done before. I got a 2019 iMac 5K i9 for home and iMac Pro at work. They are pretty similar with the iMac Pro running quieter and under load for a bit longer at higher speeds, from what I can tell in regular use. But they're not as different as you might think.
 
  • Like
Reactions: Zdigital2015
Its devs actually introduced arbitrary compute pauses in order to avoid any kind of thermal throttling. A CPU benchmark that treats the CPUs nicely is imho useless.
Then things like environment temperature and the surface under the laptop come into play. Geekbench isn't useless, just only useful for burst usage, which is still a valid real-life scenario.

My 2012 MBP used to run full blast CPU + graphics (due to PoS Google Hangouts) at min fan speed... if I put it on top of an ice pack.
 
I just want a pre-2016 keyboard and form factor, with lots of still relevant ports and a magsafe power connector. Add all the other innards in this thing to that blend, and I'd still be buying MacBook Pros. Focus on thin and pretty for MacBooks and MacBook Airs. MBP's should be highly functional and adaptable to workplace needs, along with being presentably "pretty" and highly reliable.
 
  • Like
Reactions: 0837990
Alienates loyal customers. They’re not managing step-ups well anymore. In 6 months they stepped-up graphics and then processors but charge the same amount. Makes it very difficult to plan, particularly for businesses.

How do you not see that?
What do you mean it makes it difficult to plan? Are you so concerned about always having the very best machine on the market to the point that you think everybody else should get a worse machine that Apple could actually make?

I literally couldn't care less if my MacBook Pro was the absolute fastest at any given point in time. All I care about is that it performs well, and both the 2018 MacBook Pro and the 2019 MacBook Pro do so.

However, I still don't like the Butterfly keyboard. Even if they somehow managed to actually make it not break, I still don't want it. I'm just going to wait for the next redesign, hoping desperately for a new type of keyboard. Either way I will buy it, because continuing to use this 2011 MacBook Pro is just becoming untenable, but I really, really, REALLY hope it's better than the current one.

I desperately, desperately need an upgrade. While the 2011's CPU is still reasonable for what I do, the GPU is woefully inadequate, and my e-GPU solutions are clunky and really aren't working that well on the relatively slow Thunderbolt 1 connection, and I can't install the latest version of macOS anymore. Hasn't been a problem yet, but I'm sure it will be.
 
I tested the 2018 2.9GHz 6-core MacBook Pro. Geekbench 4 Multi-Core CPU score was 23315. By my calculations, the 2019 2.4GHz 8-core MacBook Pro score of 29184 is 25% faster.

But if you pick the highest posted 2018 2.9GHz 6-core MacBook Pro Multi-Core CPU score of 26148, the 2019 is only 12% faster. See https://browser.geekbench.com/v4/cpu/13267848

A more useful shootout would be rendering and/or transcoding with Final Cut Pro, DaVinci Resolve, etc. And those tests need to last long enough to trigger any possible thermal throttling.
 
I tested the 2018 2.9GHz 6-core MacBook Pro. Geekbench 4 Multi-Core CPU score was 23315. By my calculations, the 2019 2.4GHz 8-core MacBook Pro score of 29184 is 25% faster.

But if you pick the highest posted 2018 2.9GHz 6-core MacBook Pro Multi-Core CPU score of 26148, the 2019 is only 12% faster. See https://browser.geekbench.com/v4/cpu/13267848

A more useful shootout would be rendering and/or transcoding with Final Cut Pro, DaVinci Resolve, etc. And those tests need to last long enough to trigger any possible thermal throttling.

My 2018 i9 has consistently scored about 25000 on multi core, and 5770 on single core. I think that’s mostly what most people get when they run geekbeench under proper conditions (no other apps running, no indexing, not a ton of background services like Adobe Creative Cloud etc)

https://browser.geekbench.com/v4/cpu/13249279
Single Core: 5770
Multi-Core: 24931
 
Last edited:
Now, at the WWDC tim Cook will mention the new update to the Macbook, most amazing and fastest "ever"...

Should be announced as the worse Apple product designed "EVER"

What is sad is that Tim Cook announced that this will be a great year, and so far has been extremely disappointing.

All updates, iMac, Macbook Pro, MacBook Air (except Mac mini) has been mere internal upgrades.
 
What is sad is that Tim Cook announced that this will be a great year, and so far has been extremely disappointing.

Not that I disagree. However, no surprises here, since Tim Cook always claimed to have an amazing new products pipeline ... a pipeline which proved to be yawn inspiring so far. Nothing to really mention (No. The Apple Watch is not worth mentioning in that department), now 8 years and counting.
 
  • Like
Reactions: Peperino
They’ll probably release the 16” with a redesign, OLED screen, super efficient cooling, better graphics and history’s best keyboard in like 3 months for the same amount. Seems to be the way it’s going these days.
If only. Judging by the premium on the touch bar redesign, the OLED will add +$500 over the cost of the previous model alone. My 2016 15” maxed (except for the $$$$ 4TG ssd) was the most in have spent for any Mac i’ve owned since 1991.
 
MBP with 6 core is anywhere from 10 to 30% slower than a Windows machine with the same CPU, depending on how big the other one is, under sustained load of course. In Geekbench they will look similar. The gap will be much larger with the 8 core.
You say that like it’s true. Prove it.
 
You say that like it’s true. Prove it.
It is not exactly rocket science, MBP can dissipate 55W of heat from CPU, "other brands" can dissipate more. The base 2.2GHz 6 core can pull 65W at sustained max turbo, the top 2.4GHz 8 core probably close to 130W. The machine which can discharge more heat will run at faster clock rates, simple as that. Want examples? OK, just so it happens I own both 2018 2.2 MBP and a MSI GS65 laptop with the same i7-8750h CPU. In Cinebench R15 I get 1050 on MBP, 1250 on MSI. In R20 it is 2550 on MBP and 2950 on MSI. And so on. Mind you my MSI is actually limited by max CPU clock most of the time, not thermals like MBP - so a CPU with higher max clock (or higher core count) requiring larger power will increase the gap. And while GS65 is comparable in size and weight to MBP and maxes out close to those 65W, there are monster laptops out there capable of cooling pretty much an overclocked desktop CPU and a garden variety of products in between.
 
Not that I disagree. However, no surprises here, since Tim Cook always claimed to have an amazing new products pipeline ... a pipeline which proved to be yawn inspiring so far. Nothing to really mention (No. The Apple Watch is not worth mentioning in that department), now 8 years and counting.
Watch, AirPods, iPhone X, mobile silicon, etc..


You ignoring the great products from Apple doesn’t mean they aren’t here.
[doublepost=1558959394][/doublepost]
It is not exactly rocket science, MBP can dissipate 55W of heat from CPU, "other brands" can dissipate more. The base 2.2GHz 6 core can pull 65W at sustained max turbo, the top 2.4GHz 8 core probably close to 130W. The machine which can discharge more heat will run at faster clock rates, simple as that. Want examples? OK, just so it happens I own both 2018 2.2 MBP and a MSI GS65 laptop with the same i7-8750h CPU. In Cinebench R15 I get 1050 on MBP, 1250 on MSI. In R20 it is 2550 on MBP and 2950 on MSI. And so on. Mind you my MSI is actually limited by max CPU clock most of the time, not thermals like MBP - so a CPU with higher max clock (or higher core count) requiring larger power will increase the gap. And while GS65 is comparable in size and weight to MBP and maxes out close to those 65W, there are monster laptops out there capable of cooling pretty much an overclocked desktop CPU and a garden variety of products in between.
That’s all a fancy way of saying it depends on the work load. That doesn’t mean it’s slower as a matter of fact.
 
Watch, AirPods, iPhone X, mobile silicon, etc..
You ignoring the great products from Apple doesn’t mean they aren’t here..

Not ignoring. The Watch did not exactly punch a dent into the universe. Airpods... an accessory. iPhone an evolution, not new, not that exciting.
Admittedly, the Ax Chips are great. But again Mr. Cook inherited it and so far no new product has emerged from it (iPad/iPhone have been around for quite some time, in case you missed )
 
Not ignoring. The Watch did not exactly punch a dent into the universe. Airpods... an accessory. iPhone an evolution, not new, not that exciting.
Admittedly, the Ax Chips are great. But again Mr. Cook inherited it and so far no new product has emerged from it (iPad/iPhone have been around for quite some time, in case you missed )
Look at the sales for wearables...not exactly a small business anymore. iPhone X propelled Apple to $166B in iPhone revenue in a single year. That’s over 3X what Jobs did with his iPhone strategy. Do you know how insane $166B is for a single product? That doesn’t just happen...you have to execute and create a value story. Cook did that with X and pushed ASPs to $765

You poo pooing the wearables market just ignores the facts. It’s a HUGE business growing at 30% and $20B in sales, making it a Fortune 200 company on its own.

Even iPad returning to growth shouldn’t be ignored. The management of Apple has been superb.
 
If only. Judging by the premium on the touch bar redesign, the OLED will add +$500 over the cost of the previous model alone. My 2016 15” maxed (except for the $$$$ 4TG ssd) was the most in have spent for any Mac i’ve owned since 1991.

I was looking at specs on the newest Razor Blade 15. I'm drooling over the OLED screen and RTX 2080 GPU options. Wish Apple would offer those options along with better cooling even if it meant a thicker form factor.
 
  • Like
Reactions: ImaxGuy
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.