Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
OK, se I've been using my 2.3GHz 15" rMBP for a few hours and here are a few impressions

• Battery life is fantastic. On my 2012 15" I was getting at most around 4 to 5 hours out of it. This one is quoting me around 9 hours which is great.
• When using the Iris HD graphics (haven't tried the 750M yet) this machine is smooth. There's nothing anywhere near the lag I got on my 2012 machine.
• Unfortunately the screen temperature is not uniform across the display. On the left hand side it looks yellowish and almost dirty, whereas on the right it's more neutral. Plenty of people have gone through this kind of thing before with Retina MacBooks so I won't labour the point, but it's bad enough that I'm going to return it. I haven't decided if I'll take a replacement or a refund yet, but I don't think there's any point in waiting out a "fix" for this at the factory, experience would suggest that it seems to just be the luck of the draw.

Here's a photo:
Image

+1

I don't know what to do. I need a new laptop but at the same time...that screen. I keep trying to say to myself i'll mostly be looking at my Cinema Display but that isn't a good justification. Why can't panel manufacturers just get this right.
 
Returning the Haswell 15" [ 2.6 / 16 / 512 all Samsung ] because it's [ for my needs ] no better than my 2012 one [ specs in sig ]

I have zero lag on my 2012 and it's just overall amazing and still very new! Battery life is great, but it's 9/10 times plugged in so that advantage [ even though trivial ] is moot.

I believe the Haswell release might be good jump if you have a pre-Sandy Bridge unit, or must squeeze every little bit from the CPU, GPU and SSD.

Haswell has worked great for the new MBA and probably the new 13" rMBP, but I'm not sure how significant the boost is for the 15" rMBP :)
 
I must say, hats off to Apple: their pricing structure this time around is an effing masterstroke. I can just see Tim stroking his chin and muttering, "Checkmate," as he signs off on the final prices. "That'll be $2600, please and thank you."

Yeah, I was commenting about this after the keynote back when everyone was screaming in the "Waiting for Haswell" thread about how the pricing didn't make sense. From a business perspective, it makes perfect sense. It's freaking brilliant.
 
+1

I don't know what to do. I need a new laptop but at the same time...that screen. I keep trying to say to myself i'll mostly be looking at my Cinema Display but that isn't a good justification. Why can't panel manufacturers just get this right.

Did you get a similar screen?
 
Returning the Haswell 15" [ 2.6 / 16 / 512 all Samsung ] because it's [ for my needs ] no better than my 2012 one [ specs in sig ]

I have zero lag on my 2012 and it's just overall amazing and still very new! Battery life is great, but it's 9/10 times plugged in so that advantage [ even though trivial ] is moot.

I believe the Haswell release might be good jump if you have a pre-Sandy Bridge unit, or must squeeze every little bit from the CPU, GPU and SSD.

Haswell has worked great for the new MBA and probably the new 13" rMBP, but I'm not sure how significant the boost is for the 15" rMBP :)

+1
You put it aptly.
My Haswell is going back Sunday, sticking with 2012.
 
I guess you didn't see my post, i am getting a lot more then a 1 hour boost, I have both. I get 10+ hours on full charge.

Ah, sorry, yeah I missed that. Interesting. See below...

• Battery life is fantastic. On my 2012 15" I was getting at most around 4 to 5 hours out of it. This one is quoting me around 9 hours which is great.
• When using the Iris HD graphics (haven't tried the 750M yet) this machine is smooth. There's nothing anywhere near the lag I got on my 2012 machine.
Had you put Mavericks on the 2012 machine? I've been running it for a while, and the difference was night-and-day, with no lag in the usual tasks.

My question to both of you is the following: I'd be really curious about a couple things:
• Does CoconutBattery show any difference in capacity for the two batteries? Either out of the gate, or diminished due to time and charge cycles? I'd like to eliminate that as a possible explanation for differences.
• Otherwise, any chance you could rig up an experiment of sorts with the two models? prestong, it sounds like maybe you can't, since I'm gathering you don't have the old 2012 model anymore, but it sure would be nice to, say, set some webpages to auto refresh, or have some videos running on loops, or something like that.

I'm totally perplexed as to why, if what you guys are seeing is accurate, Apple's suddenly going conservative with their battery marketing numbers. And, if what you guys are seeing is right, it suddenly shifts my vote back in favor of the Haswell models over closeout/refurb 2012 and early 2013 Ivy Bridge models.
 
My brand new Haswell 15" rMBP (with the GT 750m) is quoting 9.5 hours remaining at half the maximum brightness with 95% battery remaining. This is browsing the internet with Bluetooth and WiFi on.

It is ridiculously fast. I don't have a 2012 rMBP to compare it too but will have access to one in the next couple of days. I'll do a few tests and let you guys know how they stack up.
 
Ah, sorry, yeah I missed that. Interesting. See below...


Had you put Mavericks on the 2012 machine? I've been running it for a while, and the difference was night-and-day, with no lag in the usual tasks.

My question to both of you is the following: I'd be really curious about a couple things:
• Does CoconutBattery show any difference in capacity for the two batteries? Either out of the gate, or diminished due to time and charge cycles? I'd like to eliminate that as a possible explanation for differences.
• Otherwise, any chance you could rig up an experiment of sorts with the two models? prestong, it sounds like maybe you can't, since I'm gathering you don't have the old 2012 model anymore, but it sure would be nice to, say, set some webpages to auto refresh, or have some videos running on loops, or something like that.

I'm totally perplexed as to why, if what you guys are seeing is accurate, Apple's suddenly going conservative with their battery marketing numbers. And, if what you guys are seeing is right, it suddenly shifts my vote back in favor of the Haswell models over closeout/refurb 2012 and early 2013 Ivy Bridge models.

Coconut Battery on my 2012 model reports that its capacity is at 89% after 1 year and around 350 cycles, whereas I assume the Haswell model is at 100%.

I'm running Mavericks on the 2012 machine and perhaps it was a bit unfair to describe it as "lag" - everything on this machine is perfectly responsive, but the 2013 model is still noticeably smoother. When switching between spaces, for example, the switch happens in the same amount of time - the 2013 machine is just able to do it with more FPS and so it feels sleeker.

Unfortunately I can't do an experiment because it's all boxed up and ready to go back - but my guess as to why its getting more time than advertised is because Apple's always been conservative about their battery life estimates, and I think I might be right in saying that they said they made their battery life tests tougher when they launched the Haswell Airs, so a change in testing methodology may explain part of the discrepancy.
 
+1
....
My Haswell is going back Sunday, sticking with 2012.

Likewise :D

I hate being impulsive when it comes to tech.. :eek: So gullible...

----------

....

I'm running Mavericks on the 2012 machine and perhaps it was a bit unfair to describe it as "lag" - everything on this machine is perfectly responsive, but the 2013 model is still noticeably smoother. When switching between spaces, for example, the switch happens in the same amount of time - the 2013 machine is just able to do it with more FPS and so it feels sleeker.

....

Not questioning your claim, but I see no difference and I'm testing both units right now [ the Haswell is 2.6 / 16 / 512, Samsung ]

I just cannot see where/how it's smoother over the 2012 model. I'm switching between spaces, maximizing App Store window, scrolling FB and comparing PhotoShop, Aperture, and X-Code [ and some other compilers ]... PS opens maybe a fraction of a second faster... that's all ..

I cannot tell a difference except in boot-up speeds and Disk read/write tests...

Running iStatPro on both, the Haswell shows slightly higher temps over Ivy on the dedicated GPU.
Intel Power Gadget does not show any CPU throttling in either one under heavy load [ stressing with decoding/encoding dvd ; running 2 VMs [ Linux and SQL ], MS outlook and excel running in the background, about 12 tabs open on Chrome and 3-4 on Safari, and doing a TM backup at the same time...]
 
Last edited:
Did you upgrade your old retina to mavericks, it gets ride of allot of the lag issues.

OK, se I've been using my 2.3GHz 15" rMBP for a few hours and here are a few impressions

• Battery life is fantastic. On my 2012 15" I was getting at most around 4 to 5 hours out of it. This one is quoting me around 9 hours which is great.
• When using the Iris HD graphics (haven't tried the 750M yet) this machine is smooth. There's nothing anywhere near the lag I got on my 2012 machine.
• Unfortunately the screen temperature is not uniform across the display. On the left hand side it looks yellowish and almost dirty, whereas on the right it's more neutral. Plenty of people have gone through this kind of thing before with Retina MacBooks so I won't labour the point, but it's bad enough that I'm going to return it. I haven't decided if I'll take a replacement or a refund yet, but I don't think there's any point in waiting out a "fix" for this at the factory, experience would suggest that it seems to just be the luck of the draw.

Here's a photo:
Image


----------

agreed, I also don't notice any difference in real world usage.

Likewise :D

I hate being impulsive when it comes to tech.. :eek: So gullible...

----------



Not questioning your claim, but I see no difference and I'm testing both units right now [ the Haswell is 2.6 / 16 / 512, Samsung ]

I just cannot see where/how it's smoother over the 2012 model. I'm switching between spaces, maximizing App Store window, scrolling FB and comparing PhotoShop, Aperture, and X-Code [ and some other compilers ]... PS opens maybe a fraction of a second faster... that's all ..

I cannot tell a difference except in boot-up speeds and Disk read/write tests...

Running iStatPro on both, the Haswell shows slightly higher temps over Ivy on the dedicated GPU.
Intel Power Gadget does not show any CPU throttling in either one under heavy load [ stressing with decoding/encoding dvd ; running 2 VMs [ Linux and SQL ], MS outlook and excel running in the background, about 12 tabs open on Chrome and 3-4 on Safari, and doing a TM backup at the same time...]


----------

My thoughts exactly

Returning the Haswell 15" [ 2.6 / 16 / 512 all Samsung ] because it's [ for my needs ] no better than my 2012 one [ specs in sig ]

I have zero lag on my 2012 and it's just overall amazing and still very new! Battery life is great, but it's 9/10 times plugged in so that advantage [ even though trivial ] is moot.

I believe the Haswell release might be good jump if you have a pre-Sandy Bridge unit, or must squeeze every little bit from the CPU, GPU and SSD.

Haswell has worked great for the new MBA and probably the new 13" rMBP, but I'm not sure how significant the boost is for the 15" rMBP :)
 
Well after a day of burn in I am pleased to announce my battery life is still great. 86% left and it predicts 10 hours, very happy I did the upgrade, check out the screenshot..
 

Attachments

  • Screen Shot 2013-10-24 at 4.13.21 PM.jpg
    Screen Shot 2013-10-24 at 4.13.21 PM.jpg
    979.6 KB · Views: 346
It's no different for hardcore pro users, either. The only people who should really notice any performance difference are:
1) People who thrash their disk, and thus might benefit from the PCIe storage
2) ThunderBolt 2, which again only matters for a subset of users
3) People who do OpenCL-intensive stuff
4) People using an 802.11ac access point

People who do heavy OpenGL stuff will often be better with the older base model versus the new base model. And CPU wise, model-to-model, it's about the same.

This really is a price drop to go with a terribly ho-hum upgrade.

Pretty much spot on. Improvement is minor at best, but a lot of that is Intel's fault.
 
I cannot tell a difference except in boot-up speeds and Disk read/write tests...

How noticeable a difference is it for boot-up? I'm thinking of that somewhat as a proxy for virtual machine performance, since that often tends to be I/O bound (an under-appreciated fact by many).

----------

Well after a day of burn in I am pleased to announce my battery life is still great. 86% left and it predicts 10 hours, very happy I did the upgrade, check out the screenshot..

I commend you on the wisdom of including a john123 screenshot. :D
 
Its official Apple's version of the 750M runs at 926MHz. I used a CUDA sample program provided by Nvidia to determine that.

I guess it might be under clocked. For reference, 650M runs at 900 Mhz on the rmbp.
 

Attachments

  • Screen Shot 2013-10-24 at 7.19.45 PM.png
    Screen Shot 2013-10-24 at 7.19.45 PM.png
    332 KB · Views: 286
Its official Apple's version of the 750M runs at 926MHz. I used a CUDA sample program provided by Nvidia to determine that.

I guess it might be under clocked. For reference, 650M runs at 900 Mhz on the rmbp.

meaning it wont get that hot+plus a little less battery drain right?

if you were an architect like me, yould you go for the dgpu or the igpu mac?
 
Its official Apple's version of the 750M runs at 926MHz. I used a CUDA sample program provided by Nvidia to determine that.

I guess it might be under clocked. For reference, 650M runs at 900 Mhz on the rmbp.

So, there will be no difference between the 650m and the 750m. Apple is basically lying saying that they put a 750m in as far as I am concerned. I mean at least it is deceptive, because one would expect to at least get the manufacturer specs. I'm very disappointed with apple.
 
What programs do you use?

-currently im using:
cinema 4d (right now main my modelling,viz and rendering app), adobe package, archicad


soon start with rhino and autocad (my current 2206 model 32 bit wont run these)


also planning to start with:

-Windows with:
revit, 3ds and rhino and autocad architecture edition.

im a little dependent to be able to do all of that since im freelance and work in teams using different applications but i try to be on OSX as much as possible.
 
Agreed.

That why didn't mention the 750m in any of their marketing benchmarks.

You won't see any real improvements till next year in terms of discrete gpu.

They will probably do the same strategy, make the base without the discrete and charge a ton for ones with a discrete.

So, there will be no difference between the 650m and the 750m. Apple is basically lying saying that they put a 750m in as far as I am concerned. I mean at least it is deceptive, because one would expect to at least get the manufacturer specs. I'm very disappointed with apple.


----------

I would go with a discrete gpu. I am not 100% sure but your apps could use OpenGL heavily.

I would go with the 2.3 Ghz Haswell, if you don't want to look at refurb Ivy Bridge's. Could get a bargain on the old Ivy Bridge's you gotta look around.

I would only get used Ivy Bridge if it is under warranty (strongly recommend getting apple care with your discover or amex to extend the warranty). I had to replace my Samsung Screen on mine, luckily I was under warranty. The screen cost $500+.




-currently im using:
cinema 4d (right now main my modelling,viz and rendering app), adobe package, archicad


soon start with rhino and autocad (my current 2206 model 32 bit wont run these)


also planning to start with:

-Windows with:
revit, 3ds and rhino and autocad architecture edition.

im a little dependent to be able to do all of that since im freelance and work in teams using different applications but i try to be on OSX as much as possible.
 
So, there will be no difference between the 650m and the 750m. Apple is basically lying saying that they put a 750m in as far as I am concerned. I mean at least it is deceptive, because one would expect to at least get the manufacturer specs. I'm very disappointed with apple.

Yes apple is deceptive because they they mentioned that the new rMBP will have the 750m with 2GB of ram and you assumed it would be miles ahead of the 650m.

In all the benchmarks and test the 750m with the 2GB performs slightly better, overall less than 5% improvement in most areas.
 
How noticeable a difference is it for boot-up? I'm thinking of that somewhat as a proxy for virtual machine performance, since that often tends to be I/O bound (an under-appreciated fact by many).

----------



I commend you on the wisdom of including a john123 screenshot. :D


About 2.. maybe 3 seconds faster on the Haswell
 
Agreed.

That why didn't mention the 750m in any of their marketing benchmarks.

You won't see any real improvements till next year in terms of discrete gpu.

They will probably do the same strategy, make the base without the discrete and charge a ton for ones with a discrete.



----------

I would go with a discrete gpu. I am not 100% sure but your apps could use OpenGL heavily.

I would go with the 2.3 Ghz Haswell, if you don't want to look at refurb Ivy Bridge's. Could get a bargain on the old Ivy Bridge's you gotta look around.

i have not interest in refurb and have had bad experience in 2012 with two MBPrs.

are you saying that 750 is always faster than iris pro when it comes to openGL?
 
Yes apple is deceptive because they they mentioned that the new rMBP will have the 750m with 2GB of ram and you assumed it would be miles ahead of the 650m.

In all the benchmarks and test the 750m with the 2GB performs slightly better, overall less than 5% improvement in most areas.

It's deceptive because as far as I can tell the 750m is 967mhz factory spec but we are getting less than this at about 920mhz. If apple is going to make the 750m worse than factory spec, they should be honest about it. If it just says 750m you should at least expect factory specs that's what I think. Right???
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.