The fans can't compensate for heat soak the just move it a bit. There has to be enough surface area and air movement.
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.
2016 nMP
- Thread starter ManuelGomes
- Start date
-
- Tags
- nmp
- Sort by reaction score
- Status
No you're oversimplifying, computer cooling is not as cooling in a car, heat soak is not as simple, first require the thermal core to reach a "soak" temperature, in case it operated at full TDP previous shutdown what only happens is the thermal core normalize its temperature (the temperature along the core is uniform), this doesn't happen as quick, and before the thermal core normalize its temperatures there is the convection natural cool down, depend on your room temperature will take away 1/3 to the half the thermal core temperature before there is any probability to overheat component to reach a "heat soak"condition.
and with to mention computers (silicon) cooling is in no way similar to car engine block (steel) cooling.
Last chance for Apple this June.
I've been on a Mac for over 20 years but unless they come to their senses I will probably have no choice but to go Win10 and believe me I am not looking forward to that.
Historically speaking Apple has never been on the bleeding edge of any of this, but the current situation is truly pathetic.
Last edited:
I need a new workstation this summer as well, a single socket 8c Xeon v4 with 64GB of ECC memory. If there is no new nMP, I guess it's a big and loud HP Z440.
If the hardware isn't ready for release until July, then release in July. There is zero good reason in Apple's modus operandi to talk about stuff substantially before you ship it unless one of two things.
i. about to kill it ( e.g., the WWDC 2013 ... which is more about "we are going to stop selling towers, if you want one best buy it in next couple of months")
ii. is a sustantially new product. (i.e., iPhone, iPad product line (not individual models), iWatch , etc.).
( 2013 Mac Pro is a bit of a corner case of that too. )
Computex ( a computer show in Taiwan that typically falls are week or so before WWDC ) is typically where Intel has released some processor upgrades aimed at the laptop line. It has made sense to line up the MacBook ( pro and not) with that release. But the driver there is Computex and Intel not "we need fodder for WWDC".
that is somewhat past. Intel release schedule isn't as synched up to Computex and is off annual "tick/tock". Furthermore, the context that WWDC has to cover ( Apple's OS and associated libraries) has gotten bigger. Apple doesn't have one or two OS to cover any more. They have four. It has at least doubled. There is more they should be covering in the keynote about the conference at hand.
The keynote dog-and-pony show isn't primarily aimed at developers. That is somewhat of a problem. They have to tread a line between being informative about what core elements of the conference are actually focused on and being a sales-pitch to broad consumer advertising ( "You should download the OS major upgrade as fast as you can because this new stuff is super cool." ).
Any new MacBook Pro is likely primarily just going to be "thinner" and incrementally better. Mac Pro that is even less as it probably will just be incrementally better. ( Personally, I think it should be a bit, ~1", taller and wider , but I doubt sensible engineering concerns will pass muster with Ive's groupthink design bureau. )
As far as external displays go .... Apple exited the display business years ago. Maybe a new docking station, but a display is likely up there with waiting for a unicorn to be presented.
Are we talking about cars?
To cool computers you need surface area and air movement its unbelievably simple.
I think Apple may simplify it's Macbook line, leaving it on 12,14 & 16 inch models having no dGPU But TB3 and slimmer than a retina Macbook Pro, and updating the current rMBP 15 to Skylake Xeon and Polaris dGPU & TB3 as the only pro MacBook, and the MacBook Air has to die.
Those on the previous rMBP 13 will move to top line rMB 14 and those on the base rMBP 15 will move to the rMB 16
The pro is as wide as it is to fit into 4U, so that is not negotiable sadly. Going taller seems nice but has some non obvious issues.
With the way things are going an nMP update with gddrx5 looks like a realistic possibility; depends on how long they wait to update it.
Metal is about having a gpu management framework that can integrate with the higher level ui frameworks in a finer grained way than OpenGL. In other words, the goal is to be able to attach custom shader snippets to individual ui widgets, everything else is just necessity once you take over the lowest level gpu stuff.
Not how metal is pitched to people but why it exists.
With the way things are going an nMP update with gddrx5 looks like a realistic possibility; depends on how long they wait to update it.
Metal is about having a gpu management framework that can integrate with the higher level ui frameworks in a finer grained way than OpenGL. In other words, the goal is to be able to attach custom shader snippets to individual ui widgets, everything else is just necessity once you take over the lowest level gpu stuff.
Not how metal is pitched to people but why it exists.
I've been eyeballing the HP Z-series over the last few years, and while I think it could be a good fit, weekly frustrations with my printer never fail to make me vow to never buy any HP product ever again.
Add frequent frustrations with my wife's Windows laptop (Dell XPS13, widely hailed as the best in its class), and I'm thinking the grass may not really be all that green over yonder.
Add frequent frustrations with my wife's Windows laptop (Dell XPS13, widely hailed as the best in its class), and I'm thinking the grass may not really be all that green over yonder.
My last windows based laptop was an Dell, a nightmare I'll never repeat is to buy Dell, despite favorable reviews, Dell is an no-no, the few times I considered a workstation I'll build mine on an gigabyte or supermicro motherboard an the case where it can fit (if any, wall mounted setups looks amazing).
On notebook I'll consider lenovo or MSI over HP, simple they keyboards are the best MSI also having mechanicals ones.
Thanks Steve we have Apple
Ya know. If Apple did indeed buy/build a place to make the Mac Pro they're not gonna just get up and walk away.
There will be a new Mac Pro. When ? I have no idea. But there will be one.
There will be a new Mac Pro. When ? I have no idea. But there will be one.
It's isn't greener, but it grows a lot faster....
Bleeding edge hardware in a workstation? No thanks.
That leaves out a lot of HP & Dell workstations.
That leaves out most Quattro & Firepro workstation graphic cards.
Premium brand....premium prices. No different than most premium brands.
Ok, so they assemble them in America. In that case, no one actually builds their own computer...they assemble it too.
Last edited:
They didn't buy/build a place. Mac Pro is built in contractor space just like the vast majority of Apple products. Apple pays them to "keep the lights on" and likely directly paid for (and perhaps "lease" or perhaps station) some of the more expensive jigs, but :
"..Apple's Mac Pro manufacturing facility is run by Flextronics as part of an initiative to bring manufacturing of some Apple products back to the United States. ..."
https://www.macrumors.com/2014/06/06/cook-visits-mac-pro-factory/
'Run by' is usually warped up to the "hype" of this being an "Apple factory". It isn't. If follow the "run by" iink in the macrumors article ... you find ....
"... That will see the company employing as many as 1,700 workers to build a "next generation desktop computer" at the company's facilities in Austin, Texas. While Flextronics will not reveal the identity of the computer or the company behind it, speculation naturally points toward Apple's radically redesigned Mac Pro ... "
https://www.macrumors.com/2013/10/1...neration-desktop-computer-likely-new-mac-pro/
So "run by" yeah as in they, Flextronics, own it. Apple is a customer. The typical spin is that Apple built/bought/runs .... they didn't and don't. Any more they "ran' that Sapphire plant that went bust. Apple only bought that after it went bust and will make nothing there but new digital bits inside a data center.
There is some piecemeal work that Apple does on iMacs in CA, but mainly in Ireland ( for EU markets ). Other than that they don't make any Macs anymore than they make iOS devices.
[doublepost=1459631108][/doublepost]
If your fans are not directly coupled to 'air movement' then have some major defects in fan design.
This is a more of problem when have a hodge-podge of fans in relatively chaotic fashion. Non chaotic air flow over the surface area will work better.
What Apple needs is something that takes the cooling requests from each of the three sources and applies that to the one fan. Suboptimally can fake that with another set of sensors that are centrally link than generate the indirect feedback but.
Technically can get to soak if keep pushing past "max fan" air flow and sustained heat input. However, reasonable firmware settings should be setting "self throttle" feedback back to the heat sources before hit that mark though.
[doublepost=1459633115][/doublepost]
Thermal core isn't "perfect" as much as well matched to the context in which all three heat sources are usually not 100% fully discharging all at the same time. The GPUs and the GDDR5 are as another post highlighted are likely the bigger contributors. If they "start/stop" on a regular basis ( e.g., no OpenCL batch job at the moment. Or the user pauses to look at something ) then that assumption is a better fit. A perfect fit likely a bit over the top characterization.
"Improve" fins means more surface area. An incrementally taller (and wider) core would do just that. It just needs a incremental buffer for "max expected" input from the sources ( which are bounded by the limited 450W power supply and the other other loads on that power supply which are not going to directly feed into the core.
The current fan doesn't even use most of the 6" diameter to move air. If Apple actually added more of the diameter to moving air it would not be hard to move substantially more at exactly the same noise levels.
Sadly I suspect those engineering detached insights seem to an infliction the Apple design team suffers from also. A narrower footprint than the Mac Mini isn't particularly necessary. The case lock is a kludge. The whole Wifi subsystem could hardly be placed into a better position to be "cooked".
Last edited:
The whole design is relatively rack hostile so why would rack measurement units be a completely non negotiable design constraint???? The previous Mac Pro was 8.1 between 4U ( 7" ) and 5U ( 8.75" ) and the world didn't end. Given you'll need a cradle to hold the round object to fit in a square hole the 0.4" seems a bit of a tight squeeze for a strict 4U fitting.
Heavier and slightly more expensive components? I don't really see the 'show stopper' issue. Perhaps a put more money in Apple's pocket issue, but technical one? Yeah it is quite opaque as to what these might be.
People complaining that the 6,1 has only a 450W PSU (rather than the ~1000W one in the cMP) probably overlook the fact that the newer model no longer needs to have reserve power for possible internal expansions. The cMP needed all that extra juice just in case someone would max out the PCIe slots and HDD bays.
With the 6,1 that is no longer the case, so Apple could just outfit it with the power it needs, rather than what end-users might potentially need.
I know there have been some anecdotes about Apple underclocking the GPU's and it was suggested that this was because of the modest PSU. I don't know about that, but it doesn't make a whole lot of sense to me; they started with a clean slate, knew exactly what they were going to put in there, so why skimp on power? The difference between a 450W or 550W PSU isn't that big.
With the 6,1 that is no longer the case, so Apple could just outfit it with the power it needs, rather than what end-users might potentially need.
I know there have been some anecdotes about Apple underclocking the GPU's and it was suggested that this was because of the modest PSU. I don't know about that, but it doesn't make a whole lot of sense to me; they started with a clean slate, knew exactly what they were going to put in there, so why skimp on power? The difference between a 450W or 550W PSU isn't that big.
My only complaint is that the design itself limited them in the graphics cards they could use. I think a big reason we haven't seen a new mac pro is that AMD's Hawaii GPUs used significantly more power than Tahiti, so while you could stick 2 of them in the tube and down clock them the increase in performance would be minimal. Had they made the tube slightly larger they could have bumped up the system power consumption to 550 W and used 2 175 W GPUs instead of 2 125 W GPUs. Making the chassis slightly larger would increase the cooling system size and maintain quietness.
Then again, AMD has already said that it sees small die GPUs and multi GPU setups as the way into the future. I wonder if Apple designed the mac pro with this in mind and the only catch was that AMD and NVIDIA ended up stuck on 28 nm one generation longer than they intended. It will be interesting this week to see if Nvidia gives any hints about how big (and power hungry) pascal is to demonstrate whether big die GPUs are coming any time soon to the 14/16 nm generation.
Wauw, 1 game
Will it be released 1 year after the big release on console / pc ? just teasing you.. games are simple not on this platform. The 99% of all the great games is not here.. and we know that it will be so for a long time
But keep in mind that heat transfer from the cooler to the air works better with turbulent flow that with laminar flow. While the fans should be placed reasonably to generate the air flow, too much aerodynamic optimization could lead to suboptimal heat transfer.
- Status