The last computer game I played was Colossal Cave Adventure, back when there were only about 600 computers connected to the Internet (then called ARPAnet) so I'm the wrong person to ask about gaming (which means I'm not a "Pro" user according to some here despite that I make my living with my computer). So, read the reviews, because I doubt any participant in this thread has actually experienced Iris Pro graphics yet.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Waiting for Haswell MBP Mega Thread
- Thread starter thadoggfather
- Start date
- Sort by reaction score
- Status
The last computer game I played was Colossal Cave Adventure, back when there were only about 600 computers connected to the Internet (then called ARPAnet) so I'm the wrong person to ask about gaming (which means I'm not a "Pro" user according to some here despite that I make my living with my computer). So, read the reviews, because I doubt any participant in this thread has actually experienced Iris Pro graphics yet.
If it's true that Apple's receiving custom chips with more powerful dGPU, wont't that cause driver problems if you would like to install Windows?
Edit: https://www.macrumors.com/2013/07/26/intel-to-supply-apple-with-special-high-end-haswell-processors-for-macbook-pro/ (link to the article)
Edit: https://www.macrumors.com/2013/07/26/intel-to-supply-apple-with-special-high-end-haswell-processors-for-macbook-pro/ (link to the article)
Last edited:
I've been reading every single post since 15 July, and all I mean is that after more than one year without new releases it's time for a serious improvement, and Iris Pro just isn't.
And if i'm not wrong, you also stated the economic reasons to launch haswell rmbp on 23 July
I've been reading every single post since 15 July, and all I mean is that after more than one year without new releases it's time for a serious improvement, and Iris Pro just isn't.
And if i'm not wrong, you also stated the economic reasons to launch haswell rmbp on 23 July
New releases do not depend on Apple but Intel...
If you really have read all of those posts, then you've read the roughly 10 reasons why that isn't a compelling argument from a business sense. Your post basically is a "Wahhh, I want it!" Haswell is a serious improvement—in battery life. Not every component has to be upgraded. And we've seen regressions before.
I really wish we could make a wager on this. I would make so much money.
And if i'm not wrong, you also stated the economic reasons to launch haswell rmbp on 23 July
I did. You'll also note the important qualifier in my post—that I'm not privy to internal knowledge of things like supply chain constraints, the R&D cycle, holiday sales forecasts as a function of product newness, etc.. Those include things like the alleged story about getting higher binned iGPUs from Intel. It's a no-brainer that if those sorts of things weren't issues, we obviously would already have seen new models.
It's both, of course. It's hard to tell what the hold up was this time. Availability of the 2.4Ghz chips may have been part of the issue, or it could be the whole better iGPU rumor, or it could be some other architectural change altogether. I think we'll have a better idea when the models come outbased on what changes there arewhat the likely reasons for the delay were. Either way, it's ludicrous to think that in a perfect world where there are no constraints that Apple would have wanted to hold off releasing new models. Doing so only can only reduce profits.
That's another reason why the 750m is probably not gonna be in the haswell rMBP, or a wildcard at best. Isn't it just an overclocked 650m? Even if the next rMBP were to be a magical beast of power with an Iris Pro 5200 AND dGPU, I wouldn't think the dGPU would be one at the end of it's lifecycle?
So we'll basically get an improvement in battery life (about 1 or 2 more hours as commented) and a regression in graphics performance after one and a half year.
I expect then a considerable drop in prices plus the inclusion of thunderbolt 2, otherwise I don't know why are we waiting until its release instead of buying the current model. I think I'm not the only one expecting some extra features...
Anyway, all your comments are well stated indeed.
That's what's intriguing me til the next keynote. If the regression is going to be so drastic, what will be the shining star of the Haswell rMBP? Here's to hoping that it'll be more than marketing spin (which in all honesty is all a company sometimes needs).
There is not going to be drastic regression for anyone and there won't be any significant regression for anyone except gamers. 802.11ac will be a nice advance for almost everyone, as will generally better all around performance, and much faster SSDs.
It's probably gonna be like the WWDC was in 2012 for the cMBP. "Oh, we popped Haswell in there. Good battery life. Faster wifi. Faster flash. Oh no dGPU btw. Same price. K now let's talk about the Mac Pro!"
I don't know what you mean with over all improved performance. CPU is like 0,1% faster. The faster SSD is hard to notice. And Iris is at least 24% slower, except OpenCL. Maybe that can be noticed, maybe not. Therefore we need a real world test. But this thing already has a memory and bandwith problem. Im not to optimistic for future software/games. But its Apple, they find a way to release it.
I've been reading this forum for a few months now, and I still can't figure out where everyone got the idea that Apple is definitely dropping the dGPU. I can't see any hint of it in anything other than blind speculation. Like, not even based on some kind of factual rumor.
Reasons abound from "They wouldn't include both an Iris Pro 5200 and a dGPU, duh!" to "Oh, they have a custom Haswell chip, well that guarantees no dGPU!" to "If they're going to target 12 hours of battery life, there definitely won't be a dGPU."
I remember when the 2011 MBPs came out, people were saying the same thing about quad cores. Months of people saying "there won't be quad cores, it'll destroy the battery life, it'll be too expensive, Apple has already decided that quad core is unnecessary." And, of course, I then went and bought myself the 15" quad i7 MBP.
So, does anyone have any *actual* reason to be slapping down people who think there will still be a dGPU? Or is it just a case of "my blind prediction of change is better than your prediction of status quo."
Reasons abound from "They wouldn't include both an Iris Pro 5200 and a dGPU, duh!" to "Oh, they have a custom Haswell chip, well that guarantees no dGPU!" to "If they're going to target 12 hours of battery life, there definitely won't be a dGPU."
I remember when the 2011 MBPs came out, people were saying the same thing about quad cores. Months of people saying "there won't be quad cores, it'll destroy the battery life, it'll be too expensive, Apple has already decided that quad core is unnecessary." And, of course, I then went and bought myself the 15" quad i7 MBP.
So, does anyone have any *actual* reason to be slapping down people who think there will still be a dGPU? Or is it just a case of "my blind prediction of change is better than your prediction of status quo."
As someone with a 9400 GPU, I'd be happy to "suffer" the "downgrade" to 5200 graphics!
Then, respectfully, you haven't really been reading.
That's actually not usually cited as one of the reasons, at least by anyone with a brain. There's no chance that they'd do that combination because it makes no sense. Please tell me you do understand that point, right? But that's not related to the dGPU or no dGPU point. There's an outside chance of an HD 4600 and a dGPU, but it's not likely.
Again, not cited by rational people as a reason.
Actually, battery life is expected to be pretty comparable. The passive draw from Iris Pro is going to be around or slightly above what the HD 4000 and HD 4600 did/do. Apple has historically not used the dGPU at all in its marketing collateral when talking about battery statistics; those numbers rely on normal usage using the iGPU.
Do you really need me to go write out the whole list of reasons again? I'm almost tempted to, given that people like you keep claiming to have "read" things but either haven't actually done so, or skimmed so quickly as to not absorb the information, only so as to have a giant blob to copy-paste over and over again. I'd suggest you visit my last 20 or so posts in this thread as well as the separate dGPU thread. I agree with most (although not all) of mseth's points on the matter as well.
----------
Clock for clock, the CPU is actually quite a bit faster. The current 2.7 Ivy Bridge is on par with the leaked 2.4 Haswell part. The only thing rendering that irrelevant is that the GT3e parts come in speeds of 2.0, 2.2, and 2.4, which will mean that assuming Apple goes with the Iris Pro 5200, we won't be seeing much of a speed increase. Haswell does come in faster mobile speeds; it's just unlikely with the Iris Pro 5200 junk that we'll see them in MBPs.
Thus, the only real regression we'll see is for gamers and people who depend heavily on OpenGL stuff all the time. For everyone else, performance will stay about on par while battery life will increase. That may not be "progress" to all people, but I find it satisfactory enough. People seem to expect miracles from each new iteration and then become disappointed when they don't happen. Right now, we're in an era of evolutionary versus revolutionary improvement on these things. Folks need to temper their expectations accordingly.
I see your point john123 and it totally makes sense for Apple. In 2010 Apple was upset about the loss of nVidias northbridge for the 13" MBP and forced Intel to build acceptable iGPUs. iGPUs have for the first time middle class performance and make cards like the 730m or even the 740m obsolete. More space for battery, less parts, no gpu switching etc. We already discussed this.
But this middle class performance could only be held up for a short time. AMD will push out their new beefier laptop cards by the end of this year and nVidia in march 2014. And Intel has nothing to offer until broadwell in late 2014. IGPU sound like a good short term solution, but in the long run they totally depend on Intel. If we see a 15" rMBP with iGPU, there could be a comeback of the dGPU next year, because the competition will offer much more performance. And Apple brought some things back like the keyboard backlight in the Air.
But this middle class performance could only be held up for a short time. AMD will push out their new beefier laptop cards by the end of this year and nVidia in march 2014. And Intel has nothing to offer until broadwell in late 2014. IGPU sound like a good short term solution, but in the long run they totally depend on Intel. If we see a 15" rMBP with iGPU, there could be a comeback of the dGPU next year, because the competition will offer much more performance. And Apple brought some things back like the keyboard backlight in the Air.
Nope, never.
When it comes to technology, you'd expect the new generation to out-perform the previous generation in just about every category you throw at it. It should be an improvement in every category (whether 1% or 20%), but never a regression. The reason why people are pissed is not because it's got similar performance to the 650M (last years gen. dgpu), it's because it's going backwards. If this is one of those times where Apple is shifting to iGPU because in 3-4 years it's going to be the best thing for laptops, then this is a time when one of those very unlikely situations will actually occur in tech. You can't expect people to sit well with that, as it would have been unheard of a few years ago (obviously because intel graphics were crap - which we're on the cusp of now).
The point actually being, if this is one of those times, where they are making the change for the best, because of the future. Then they should have something to tell us. Either one of two things will happen. They will introduce the new rMBP with only intel graphics, and essentially elude the fact that in certain categories it is worse than the predecessors graphics (assuming we're doing a task that requires high end graphics).
Or they'll put a 760M in there, and there's gonna be some funky match up of the low end gets the faster CPU but slower GPU (Iris Pro), and the High end gets a slower CPU but Dedicated graphics.
They're obviously going to do the first, but it's just going to be shameful because of what I mentioned at the top. I feel like if they are going to do something like this, this year, then they should have prepared. They talked to Intel about seriously increasing the performance of their graphics several years ago. That means that at some point they were expecting this to happen. I just think that if they had all this time to think about this one possible refresh, where there's a chance to have notable drawbacks, they could have at least designed a external dGPU pack to follow up with their Notebook line ups too. If they do that when thunderbolt 2 comes out (or next years model) then good for them. But I seriously doubt it.
People will be pissed, (maybe not enough to warrant any care) but I'm just hoping that some reviews out there actually know what they are talking about, and mention that last years model actually outperforms this upcoming Haswell in a significant manner
Edit: And no, I'm not as ignorant as some people may think. I realize this issue means little to nothing as a whole. But I'm simply sticking up for the strain that apple is causing in certain people, over their own choices. They're choosing to make this decision, and I'm simply choosing to point out the consequences of their choice.
Last edited:
This is good to remember. As fun as it was, there may not be another "pull a laptop out of an interoffice mail envelope" moment for awhile.
As evolution goes, I'd be more willing to swallow an iGPU that performed no worse, in every way, than the 650m. I think what's harder for people to swallow (including myself, sometimes) is any kind of serious performance decrease with a new product, even if real-life use isn't impacted.
I think it just comes down to people hoping that Apple's business decisions keep them on the bandwagon.
And as always, real-life testing may surprise. Like I said in another thread:
When people on macrumors find out that Apple's dropping the dGPU:
And then when they actually buy it and test it with their apps and games:
In my experience, Haswell took the MBA from 5-6 hours of battery life to well over 14. What can we expect of the rMBP?
More to the point:
If we know these items we should be able to figure out the expected battery improvement fairly easily.
More to the point:
- Do we know how much power the screen draws on the 13" rMBP?
- Do we know what % of power the CPU draws on the 2012 13" rMBP?
- Do we know the difference in power draw between the CPU in the 2013 rMBP and the 2012?
If we know these items we should be able to figure out the expected battery improvement fairly easily.
Yes, I don't remember specifically and don't feel like searching but I believe it was hypothesized that it would be about 3-4 hours gain, factoring in 1 hour boost from mavericks. I think that was for the 15" though
I'm quoting two parts of your reply, because I agree with pretty much everything you wrote but especially these two parts. Is it a regression? Yup. Is it a regression they had to do? Nope. Does it matter to some people? Yup. Will it make any consequential difference when it comes to their sales? Highly doubtful.
The one ugly thing about Apple has always been that in a way, it's a monopolist. While some may argue that PCs are a commodity business, when the "Apple experience" is created in such a way as to differentiate itself from the rest of the market, it means they can get away with an awful lot. Historically, that's been super high prices (which led to their industry leading margins) and, until the Intel switch, specs that significantly lagged the PC market. Both of those have become better in recent years, but they can still get away with crapjust like this.
I've said all along I'd much rather have a 760M. But the fact that I won't get it won't stop me from buying a new laptop, and Apple knows there are way too many people just like me out there and not enough people who won't buy on principle because one item has a bit of a regression. Sucks, but oh well.
You are right we are definitely paying a lot for the better screen and the design and apple's products are overpriced when compared to PC products. But whenever anyone like yourself makes this argument they often leave out the operating system. I love the design and build quality of apple products but if I could load OSX onto any PC I bought I would definitely consider those options.
In my opinion because the operating system isn't portable it puts PCs in a different class than Macs. We can certainly compare hardwear specs and how specific programs run on each system (games, adobe, etc.) but not the entire products as a whole. I can only get OSX with a Mac and that's why I pay. It is not the same as buying an overpriced PC.
What percentage of consumers buying the laptop will know of any regression in performance? I know I certainly would but most if not of my friends and family would have no clue, that is until I told them. Maybe Apple is banking on the idea that most consumers don't compare previous models with current ones. They mostly compare one product line to another (pro vs air). I don't know if this is true but I have seen this behavior in my friends. Also, even when there is huge performance differences I have a hard time getting them to care.
Again, I'm not saying this is what's happening I'm just adding to the conversation.
- Status