16-Inch MacBook Pro Now Available With Up to 75% Faster Radeon Pro 5600M Graphics

[Hjupter Cerrud]

The more I look at this chip, the more impressed I am. I have a 10lbs plus Alienware 17 R3 with 1070 chip that puts out 5.9Tflops that draws 115 watts or so on the GPU. Fast forward to the 5600m that's putting out only 11% slower performance at around 50 watts. When compared to its contemporaries like the RTX 2060 Max Q, you're getting 15% better performance at 60% less power usage.

This chip is awesome, but costs an arm and a leg.

PS Did they fix the 20 watt open lid with 4K+ display bug in this one?

Tempting to trade in my 7 month old MBP 16....

Sadly trade in value is not that much, they are offering me $1.8k for a laptop that cost me $3.5k less than a month ago. This card made want to sell this computer but I will probably wait until it gets better = next gen to try to sell it (if I can).

Nice to see it performs a bit better than a 1080 GTX at least on the metal side

Metal Benchmarks - Geekbench

browser.geekbench.com

 

[Wahlstrm]

Looks like gaming will be a thing

 

[buzzman]

Gaming is never a thing on a 60Hz display

 

[chucker23n1]

If they haven’t gone to intel 10th gen core by now, can’t imagine they’ll do it soon.

I think they looked at Comet Lake-H, laughed, and decided not to bother. Slightly higher clock rates but only in short bursts and with no improved thermals? It's basically a wash. Not worth it to engineer.

11th-gen could turn out to be a lot more interesting. They could go with Tiger Lake-H, which will be 10nm, meaning not only much improved thermals but also a much newer microarchitecture in Ice Lake/Sunny Cove, and on top of that the Willow Cove improvements: larger L3 cache, much faster graphics, and stuff like PCIe 4. That might mean a faster SSD.

But I think they'll go with Rocket Lake-H. It merges many of Tiger Lake-H's improvements into a 14nm core, and is likely to offer more CPU cores (but weaker graphics). The reason I think so is that all 16 configurations have dedicated graphics anyway, so CPU perf matters more than iGPU perf.

Unless, of course, they go back to the pre-2016 days of a mixed setup where the $1999 price point offered just integrated graphics. Then they could offer those configurations with Tiger Lake. But I think they want to avoid that complexity.

The generation after, that's hopefully moot, and they move to Alder Lake-H with a big.LITTLE-like setup.

(I'm ignoring ARM here.)

I’d wait for WWDC just to see what the Arm situation is. Maybe that drives down prices on Intel machines, maybe they do announce some Arm machine. Who knows.

I don't see them offering an ARM-based MBP for quite a while. Or maybe they'll go with a hybrid approach for now.

I don't think they can afford to ditch Intel on the mobile high end that abruptly, due to virtualization and dual-boot. Apple being Apple, they might just do so anyway, but… I sure hope not.

 

[cmaier]

I think they looked at Comet Lake-H, laughed, and decided not to bother. Slightly higher clock rates but only in short bursts and with no improved thermals? It's basically a wash. Not worth it to engineer.

11th-gen could turn out to be a lot more interesting. They could go with Tiger Lake-H, which will be 10nm, meaning not only much improved thermals but also a much newer microarchitecture in Ice Lake/Sunny Cove, and on top of that the Willow Cove improvements: larger L3 cache, much faster graphics, and stuff like PCIe 4. That might mean a faster SSD.

But I think they'll go with Rocket Lake-H. It merges many of Tiger Lake-H's improvements into a 14nm core, and is likely to offer more CPU cores (but weaker graphics). The reason I think so is that all 16 configurations have dedicated graphics anyway, so CPU perf matters more than iGPU perf.

Unless, of course, they go back to the pre-2016 days of a mixed setup where the $1999 price point offered just integrated graphics. Then they could offer those configurations with Tiger Lake. But I think they want to avoid that complexity.

The generation after, that's hopefully moot, and they move to Alder Lake-H with a big.LITTLE-like setup.

(I'm ignoring ARM here.)



I don't see them offering an ARM-based MBP for quite a while. Or maybe they'll go with a hybrid approach for now.

I don't think they can afford to ditch Intel on the mobile high end that abruptly, due to virtualization and dual-boot. Apple being Apple, they might just do so anyway, but… I sure hope not.

I hope they do. Will be a much better machine, and I don’t need virtualization of x86. And only 2 percent of users dual-boot. More obviously virtualize.

 

[chucker23n1]

I hope they do. Will be a much better machine, and I don’t need virtualization of x86. And only 2 percent of users dual-boot. More obviously virtualize.

Well. Between lackluster macOS years and inability to do much of my work stuff, it would probably mean I’m getting a Dell. 😆

 

[Kalae]

I don't think they can afford to ditch Intel on the mobile high end that abruptly, due to virtualization and dual-boot.

I don’t think Apple is worried about losing customers who can’t live without dual booting Windows-for-x86. Apple isn’t interested in being tied down with their own legacy compatibility, much less being tied down by Microsoft’s.

 

[cmaier]

Well. Between lackluster macOS years and inability to do much of my work stuff, it would probably mean I’m getting a Dell. 😆

Yeah, some people will leave. For me, Xcode, working well with the rest of the Apple ecosystem, the way macOS supports my workflows, and the fact that these will be better machines than dell will keep me there. I use windows at work. It still sucks. Hopefully Apple gets their **** together and cleans up the macOS problems and makes catalyst more Mac-like too.

 

[chucker23n1]

I don’t think Apple is worried about losing customers who can’t live without dual booting Windows-for-x86. Apple isn’t interested in being tied down with their own legacy compatibility, much less being tied down by Microsoft’s.

Yes, that’s fair.
[automerge]1592577517[/automerge]

Hopefully Apple gets there **** together and cleans up the macOS problems and makes catalyst more Mac-like too.

So much basic stuff broken in Catalyst. Feels barely better than Electron. Let’s hope they have major improvements to announce next week.

 

[09872738]

Yeah, some people will leave. For me, Xcode, working well with the rest of the Apple ecosystem, the way macOS supports my workflows, and the fact that these will be better machines than dell will keep me there. I use windows at work. It still sucks. Hopefully Apple gets there **** together and cleans up the macOS problems and makes catalyst more Mac-like too.

Yup.
Heavily miss Nvidia though. Need CUDA for development. True predicament ...

 

[Made In Machines]

Who cares about 10th Gen?!? It’s hardly worth the trouble and expense at this point.

I would rather Apple skip 10th Gen and introduce a Tiger Lake H-Series model as soon as those are available to OEMs.

You'll probably end up with ARM Lake!

 

[Chompineer]

I think they looked at Comet Lake-H, laughed, and decided not to bother. Slightly higher clock rates but only in short bursts and with no improved thermals? It's basically a wash. Not worth it to engineer.

11th-gen could turn out to be a lot more interesting. They could go with Tiger Lake-H, which will be 10nm, meaning not only much improved thermals but also a much newer microarchitecture in Ice Lake/Sunny Cove, and on top of that the Willow Cove improvements: larger L3 cache, much faster graphics, and stuff like PCIe 4. That might mean a faster SSD.

But I think they'll go with Rocket Lake-H. It merges many of Tiger Lake-H's improvements into a 14nm core, and is likely to offer more CPU cores (but weaker graphics). The reason I think so is that all 16 configurations have dedicated graphics anyway, so CPU perf matters more than iGPU perf.

Unless, of course, they go back to the pre-2016 days of a mixed setup where the $1999 price point offered just integrated graphics. Then they could offer those configurations with Tiger Lake. But I think they want to avoid that complexity.

The generation after, that's hopefully moot, and they move to Alder Lake-H with a big.LITTLE-like setup.

(I'm ignoring ARM here.)



I don't see them offering an ARM-based MBP for quite a while. Or maybe they'll go with a hybrid approach for now.

I don't think they can afford to ditch Intel on the mobile high end that abruptly, due to virtualization and dual-boot. Apple being Apple, they might just do so anyway, but… I sure hope not.

Well, that comment aged like milk.

 

[chucker23n1]

Well, that comment aged like milk.

If you’re referring to today’s rumor, that’s for the 13-inch, which is on a completely different schedule.

It’s also only a rumor.

 

[Chompineer]

If you’re referring to today’s rumor, that’s for the 13-inch, which is on a completely different schedule.

It’s also only a rumor.

11/10 chance the next 16" will be ARM unfortunately. Or are you too young to remember the PowerPC phase-out?

No point in trying to hold out, Apple is not going to stick to something niche to satisfy a handful of users. They are going to do what is most profitable to them, which is using chips they themselves control the production of. No more being tethered to Intel.

A rumour that has been confirmed so many times over it is likely true. Apple is going ARM, there is no doubt here.

 

[chucker23n1]

11/10 chance the next 16" will be ARM unfortunately.

11/10? Based on what? Most seem to peg first ARM Macs for early 2021. Pretty decent chance there will be one more Intel 16-inch model this winter.

Do I see a chance that the next model will be ARM? Sure. But I'd say more like 4/10.

Or are you too young to remember the PowerPC phase-out?

lol, no, I'm definitely not too young. My first Mac was 68k.

A rumour that has been confirmed so many times over it is likely true. Apple is going ARM, there is no doubt here.

That's a completely different discussion.

 

[mike...]

If the 16" MBP is to get a refresh this year, yes it will be Intel. Will Apple spend the effort? You said yourself the 10th gen chips are barely any better?

 

[chucker23n1]

If the 16" MBP is to get a refresh this year, yes it will be Intel. Will Apple spend the effort? You said yourself the 10th gen chips are barely any better?

Right. Like I said, I think it's clear they're skipping 10th-gen (Comet Lake), but that doesn't mean they can't do 11th-gen. Probably Rocket Lake (14nm), possibly Tiger Lake (10nm).

They could make the late-2019 model their last Intel 16-inch ever, but that seems quite a leap to me.

 

[DoorsFreak87]

Does anyone think the 5600m configs will lose resale value quickly with the announcement of apple silicon? Would be a safer bet to buy the more affordable 5500m. Maybe a 5500m 8gb 64gb?

 

[magbarn]

Does anyone think the 5600m configs will lose resale value quickly with the announcement of apple silicon? Would be a safer bet to buy the more affordable 5500m. Maybe a 5500m 8gb 64gb?

They’re all going to lose major resale value if the G4 PowerBooks can be used as a guide which basically tanked in price when the MacBook pros came out. The 5600m might take a niche though as it will be the most powerful Intel MacBook Pro ever made with the most powerful dGPU...

The only thing that might keep the prices up is if the A14 or whatever flops and is only faster than the Intel’s chip in specialized tasks. It also remains to be seen if Apple can find a way to beat the 251 mm2 Navi 10 dGPU that is the 5600m using their own GPU designs whilst keeping in the same package with the CPU (I somehow doubt this and they may rely on a dGPU anyway.)

 

[DoorsFreak87]

They’re all going to lose major resale value if the G4 PowerBooks can be used as a guide which basically tanked in price when the MacBook pros came out. The 5600m might take a niche though as it will be the most powerful Intel MacBook Pro ever made with the most powerful dGPU...

The only thing that might keep the prices up is if the A14 or whatever flops and is only faster than the Intel’s chip in specialized tasks. It also remains to be seen if Apple can find a way to beat the 251 mm2 Navi 10 dGPU that is the 5600m using their own GPU designs whilst keeping in the same package with the CPU (I somehow doubt this and they may rely on a dGPU anyway.)

I think they are going to lose resale value as well. However, I am going to stay with 5500m 64gb as it may have better value than 5600m 32gb. I can only afford the 32gb config, and something tells me 64gb will be still be a great value for programmers and ITs. This will be the last 16inch MacBook Intel. I agree with the earlier posts here that said that the 5600m will not have the best resale value. After all, the card already cost 800 bucks.

 

[nilk]

Nope...Touch Bar is here to stay, learn to love it.

I'll never "learn to love it", but I did learn to deal with it by mostly disabling it and using 3rd party software to make Fn+number do Fn keys. And it's not "here to stay" until their entire product line can have it. Where is the external Touch Bar keyboard for iMac / Mac Pro (and for iPad wouldn't even make sense, but should be able to use an external one for consistency)? I use an iMac also and am at my desk most of the time I use my 16" MBP, and the Touch Bar is just a weird thing I have to deal with occasionally that gives me no benefit and is not consistent with most current generation Apple experience.

Seeing what Big Sur looks like now, and how a lot of the new interactions would be smooth with touch, starting to think Touch Bar is a stopgap until MacBooks have touch screens (or some sort of touch screen device that runs macOS).

 

[Matrix1776]

90%(?) of people who need/value specs don’t buy old/secondhand computers.

The only BTO option that might help resale is RAM, since that will one day be the difference between “still useful” and “completely useless”.

I’m quite sure it’s a lot easier to sell an older MBP/MBA with 8/16GB RAM today compared to one with 4GB. That was a $100-$200 well spent back in the day.

I would have to disagree. I typically buy higher specs as they have always resold easily. In 3 years it's doubtful 512GB SSD is going to be acceptable, whereas 4TB might still be great. I think graphics is really the only iffy thing, which is why I didn't buy the 5600M. Most will probably want whatever is the latest at that point, if that's a selling point.

 

[magbarn]

I would have to disagree. I typically buy higher specs as they have always resold easily. In 3 years it's doubtful 512GB SSD is going to be acceptable, whereas 4TB might still be great. I think graphics is really the only iffy thing, which is why I didn't buy the 5600M. Most will probably want whatever is the latest at that point, if that's a selling point.

You may have an easier time selling them, but you would be lucky to get $200 on the $1200 you spent on that 4TB upgrade. The dynamics also will be changed significantly due to the completely new architecture which will result in much faster depreciation than you're used to on Macs.

 

[chucker23n1]

Does anyone think the 5600m configs will lose resale value quickly with the announcement of apple silicon? Would be a safer bet to buy the more affordable 5500m. Maybe a 5500m 8gb 64gb?

The highest-end BTO options, regardless of the architecture switch, are only ever worth it if you really need the perf. They're never a particularly good value for the price.

Do you need that perf today? Then just buy it.

 

[DoorsFreak87]

True. I think potentially some ITs and programmers will be interested in any config with one of the i9's and 64gb of ram. The 5600m is really catered for creators like myself. I think the Intel 64gb MacBook pro 16 configs will have some decent value due to omission of bootcamp. If Intel Macs can last at least 5-6 more years(lets all hope), the programmers will likely search swappa and eBay for used Intel Macs in the coming years. If anything, it seems bootcamp will be missed by most of the programmer/IT community. It's not an easy time to get a mac, and my MBP early 2011 just crashed a month ago. I'm stuck with Intel for now.