Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

New MacPro predictions? Hopes? Fears?

Jarek Zabczynski said:
LOL...well Apple is expected to update Macs in October as well.
Click to expand...

Well... don't expect a complete Mac lineup refresh

Expect:

Hard launch two MacBook Pro models - 13" & 15" (may include thunderbolt 3 USB-C to 3.5mm headphone dongle in box)

AND maybe a soft launch for updated 5K iMac w/ dual GPU on a single board as "Pro" option (and Mac Pro replacement) - to be available Q1 2017



That's all we'll see, all other Macs to be unofficially discontinued (unless Tim Cook has courage to admit his plans for Apple publicly).

Focus is on pushing post PC world - iOS & vertically integrated business platform Apple has (near) complete control over. :angry:
 
Wayne Morellini said:
My heads under at the moment. In the late 80:s a research team in the UK, did an asynchronous ARM process I redesign. But as you say thenfeild was an issue. The bunch I mentioned above involves a patent portfolio that includes a vase patentent on asynchronous signaling. Their designs are async on a broad interpretation of the word. But whatever works, sort of thing.

Years back I studied Z notation at university, an eye opener to abstraction capability. By then it was being used in NASA. That Nasa thing again, the other bunch involves somebody that has licensed designs in NASA missions. My VOS was meant to be military grade, and it would be suitable for space usage. Nobody else in the class got it, and I finished well ahead. But only after I got ross river fever, and were seeing flying tea cups and saucers going through the bed room wall the fever was so much. After that z notation seemedba lot easier. :)

Last decade, I think, I came up with an idea to make an abstracted description of a program that could flexibly be compiled to target processors with high efficiency. Many things I went into to maximise performance of my virtual OS design (the abstracted description wasn't part of this). However, such a description of maths calculations provides long-term flexible viability across hardware types with relatively, hopefully, maximal performance. Sorry, my head is swimming so finding it hard to be to the point.

I'm glad about the metal like initiatives, as that is the way I was designing my VOS in the 80's I think. As I knew about the decrease in performance from abstraction layers, and there were a heap of them, and they introduced much slow down. I designed the API's to be as close driver or lower level as possible. There iprobably is never more than two layers (the GUI for instance), and when it makes sense direct to the hardware is used.
Click to expand...

Antony.

Interesting article on Milk. Seems like a part solution. Answers start in hardware, and progress through IS and to programming. One of the issues I identified with performance, is the patching mentality rather tjsn design mentality. Having extreme arrays of processors with minor memories, required me to come up with a solution to the locality and storage issues. It still pays to have enough local memory, buy as that is dynamically defined depending on the situation and code, it is hard to say, and a substantial bit of memory takes up the space of a number of processing units. So, you select how much at the start of design and use compromise systems to make the "best" out of it. I'll not get too much into it.

I forgot at the end of my previous post the finale thing the other bunch are rumored to have done in previiyse conpany, back when they were still doing high performance computing (prior to what I am mentioning they used to have the highest performance CPU on the market in the world, on a gatearray, just before or after ARM had the highest for a while). It is alledged, at that stage, that they could be designing so that every instruction based on the store data was decoding in parallel while the which instruction was being determined, shortening the time between instruction fetch and execution. With this technique I imagine they can dynamically refine it so the instruction chain acts as a dynamic mini pipeline. (Except they hated the idea of pipelines, eliminating such to greatly simplify and increase performance. The design did not need pipelines).
Their design dies not acvomodate a number f things that a desktop CPU does. They hated caching, DMA etc (the design can handle DMA like transfers via software, one of the few designs that can). However, to me every word of transfer are upto multiple cycles, and extra energy, wasted compared to a small DMA circuit. So I designed microcaching and DMA stuff. These are ways you can do high performance desktop level much quicker, as these features have to he emulated and basically get down to interpretation on complex desktop level stuff. The designs, as mentioned, are aimed at micro embedded and glue logic at very low energy per unit of performance, andcreasinable really good performance. They ate probably doing 2mw or less per Ghz unit by now, with more to come. So, 1000 cores might equal 2w by now. Now, mine might be 3mw or something, but achieve 10-100 times the raw desktop performance without much need to abstract to interpretation. So, gooing on other things Im researching or come up with, it might be 1mw per 10Ghz unit. However highest GHz is not what I am talking about here, it is highest useful low power frequency. So, it might be 600-1000mhz+. Once you press for more frequency, overclock you produce more leakerage, heat things up, and reduce the number of processors you can handle. But give me a suitable process and some liquid nitrogen, and let me design a circulation system, and see how far we can go (raw silicon works upto hubdeds of Gigaherz, but is slowed down by stages inntje design too. For instance, they had some things in the old days in the designs doing 7 or 12 GHz. But if you ran a chip like that back then, it would probably melt.
 
Daniel Reed said:
Well... don't expect a complete Mac lineup refresh

Expect:

Hard launch two MacBook Pro models - 13" & 15" (may include thunderbolt 3 USB-C to 3.5mm headphone dongle in box)

AND maybe a soft launch for updated 5K iMac w/ dual GPU on a single board as "Pro" option (and Mac Pro replacement) - to be available Q1 2017



That's all we'll see, all other Macs to be unofficially discontinued (unless Tim Cook has courage to admit his plans for Apple publicly).

Focus is on pushing post PC world - iOS & vertically integrated business platform Apple has (near) complete control over. :angry:
Click to expand...

If Apple ditches the headphone jack on laptops I'm dropping a nuke on Cupertino...
 
"If Apple ditches the headphone jack on laptops I'm dropping a nuke on Cupertino..."

I think the USB consortium is hoping to replace the transmission of analogue audio, and inch people towards USB. Intel for their part (half speculation) will be pushing a microchip solution to support this (not sure if there is an NDA around this?).

I was in two minds about this myself, as I have a plethora of wired headphones.

However ... after switching to some rather comfy wireless over-ear noise cancelling headphones ... I've switched camps.

Reasons

+) I can walk about 100ft and keep listening (and no tethered cable to weigh me down)
+) I can Listen a much lower volume levels (ie not deafen myself / very low listener fatigue) - Especially in high-ambient noise areas.
+) Can concentrate on the audio in hand (removes lots of low frequency aircraft & truck sounds)
+) If the (lithium ion) battery packs up (after 22 hours active use without charging), there is a 3.5mm cable to bridge .. but I have only used this once just to see of it worked.

Other thought about digital audio

+) I would like to see A->D moved to back from the microphone and absolute levels recorded at 32bit @ 48khz (or 96Khz).
(someone has commented previously that recording absolute levels is not a desirable requirement, and that sending this digitally is also not desirable. With little to no experience of the downside, I only see the long term benefits)
+) If several off-axis 'digital' microphone are used - they could be used remove low frequency sounds (eg aircraft) that are not desired (reducing ADR requirements).

Have we seen other Hardware go this route?
.. what about keyboards? After they solved the connectivity issues with keyboards (eg Apples and Logitechs first wireless keyboards), and the awful battery issues - there was no going back. Removing that wire means less clutter.

AJ
 
Why haven't they got an auto stereoscopic screen yet, it's easy and very old out of patent technology?

Quick, everybody think of desirable features that Mac notebooks should have, but don't, even bigger special addition.

Pro blade slice to put under the notebook:

Space for extra GPU/battery.

Bluray UHD 3D/burner.


Thin subwoofer function. Directional studio quality projected sound bar to your ears, or audience.

This would enhance a few professional or multimedia, situations. Considering that all they have to do is design a connection standard for the under blade, companies could design any blade they wanted to, and users in future and still attach it to the previous notebooks that support it.

Flip over tablet mode. Detachable tablet mode screen. Either with processor, or wireless desktop link to the notebook, or IMac, Mac Pro or PC.

Structural battery technology (the case etc can act as a battery, they even had solar rechargeable version a few decades ago).

So...who said Apple was innovative. I dare you Apple. Its not all about form, it about form of function.
 
Daniel Reed said:
AND maybe a soft launch for updated 5K iMac w/ dual GPU on a single board as "Pro" option (and Mac Pro replacement) - to be available Q1 2017
Click to expand...

Hmm...haven't heard about that one before.
 
MacOS updates often signal impending hardware compatibility.

+) Sierra : Base version has come out (compatible with iOS release features) without H/W alteration.

+) Sierra : That includes Drivers for H/W for Laptop & iMac - Is in Beta (this is my understanding). This will be used to test the new H/W.

+) Sierra : With fixes & tuning : Would have normally go through a multitude of BETA updates .. and rolled up in time for the new H/W; However, there have Not been a gazillion Beta's to incrementally iterate through these.

(semi) Conclusion : Apple have not had the Hardware,Testing resources or Integration time necessary to Tune the impending version of Sierra for the initial Hardware (GPU) refresh. With limited time, and with no immediate expectation of a MacPro or MacMini, all resources and testing would most likely go into Laptops, iMacs.

----

Random thoughts on Nvidia's need for Metal expertise. Apple maybe crafting a 'GPU Group 3' to their supported group of Xcode extensions. Perhaps AMDs GPUs will live on in Laptops ... and Imagination technologies will get displaced as the on-chip SoC (as Nvidia + NVLink have proven themselves with the Power CPU hookup). And on the 'always connected to the mains' category, Nvidia could creep into Stand alone Siri devices in the home?

(Pure speculation) With Imagination Technologies P&L looking less that rosy, perhaps the A10X on TSMC 10nm will begin the move to another GPU designer?

AJ
 
Last edited:
I do expect laptops and likely iMacs to get a refresh soon, in part to make the move to USB-TypeC ports. Laptops sales are declining, but not everyone is going to switch to a tablet.

The MacPro serves a different market segment; people with tons of data to deal with, people who do the kind of work we do (or similar) and some fixed installation use cases. This used to be called the workstation market - but the 6,1 nMP was barely a proper workstation on release (1,000 days ago) - let alone now.

Is Apple even interested in continuing to participate in the workstation market at this point? How many tasks are beyond the capabilities of a loaded laptop in 2016? How many serious hard core use cases are already using server racks/machine rooms/etc? What's left in between?

If I were Apple, and I wanted to sell into the remaining workstation market, a minor spec bump in late 2016 is hardly the path to redemption. Best case for those of us who don't want to deal with the wonderful world of Windows would be a more radical reinvention of the workstation that offers excellent performance per watt of energy combined with a modern OS that properly supports the latest graphics libraries/languages/etc.

If Apple really wanted to, based on the components hitting the market, I believe they could have a kick ass 7,1 MP shipping in 2017. Do they care?...

Cheers - #19
 
Whatever happened with WiFi direct? They had chipsets that beat Bluetooth in energy usage on a per bit basis.
 
Blair,

One question I asked myself was - why would Apple pump so much money into making the hardware for an Apple Watch, when they only sell a few million of them? It would not surprise me if the HW was made at a loss - BUT, the big bucks are made from their cut from Software sales.

With the MacPro (2013) they went to crazy lengths to design the robots, the extrusion systems, assembly lines, automation - again to build a few million machines. As Mac App purchases can happen outside the store - there is far less recurrent income generated, and viability is highly depend on CPU costs.

With the automated nMP build system, and distribution network already existent, and probably >90% of people with a nMP(2013) not desperately in need of an upgrade - perhaps the question at Fruitcamp is firstly, what to they need for people to trigger an wave of upgrades from existing owners? And secondly, what would Apple want to exist in a new workstation to compliment their future plans (before they would be willing to roll-out a replacement).

AJ
 
Could Apple make new graphic processor boards to fit in the current 2013 model? I thought that was one of the benefits historically of the Pro model vs iMac that it was field upgradable? I have both models and long time Mac user. I think the CPU is actually OK, even with video intensive projects, it rarely is fully utilized, but the graphics co-processor boards would make a world of difference if they could be replaced with faster models.
 
Jon Dishler said:
Could Apple make new graphic processor boards to fit in the current 2013 model? I thought that was one of the benefits historically of the Pro model vs iMac that it was field upgradable? I have both models and long time Mac user. I think the CPU is actually OK, even with video intensive projects, it rarely is fully utilized, but the graphics co-processor boards would make a world of difference if they could be replaced with faster models.
Click to expand...

There is nothing stopping Apple from offering GPU upgrades except apathy towards the Mac Pro user. Not enough courage I guess.
 
Jon Dishler said:
Could Apple make new graphic processor boards to fit in the current 2013 model? I thought that was one of the benefits historically of the Pro model vs iMac that it was field upgradable? I have both models and long time Mac user. I think the CPU is actually OK, even with video intensive projects, it rarely is fully utilized, but the graphics co-processor boards would make a world of difference if they could be replaced with faster models.
Click to expand...

Yes, they can and they were supposed to be upgraded, along with the "core" of the unit, when new hardware standards came along. Apple obviously hasn't put anything behind it, so we never got the upgraded parts that were promised when the designs were unveiled. It's possible we'll get something next year, but if we don't, I'd say the nMP is one and done.
 
Blair S. Paulsen said:
Is Apple even interested in continuing to participate in the workstation market at this point?
Click to expand...

Not really, expect a dual GPU iMac early next year for "Pro" users :badputer:
Apple is more interested in pushing the post-pc market and dominating with iOS and their vertically integrated platform.

Looks like Microsoft gets it, and is working on native linux binary and bash support via an OS subsystem that emulates a ubuntu kernal:
https://en.wikipedia.org/wiki/Windows_Subsystem_for_Linux
https://blogs.windows.com/buildingapps/2016/07/22/fun-with-the-windows-subsystem-for-linux/
 
I had to visit the Genius Bar recently at a flagship store. 3 I mac's and 1 lone nMP. The rest...mobile devices. Left the store a bit ambivalent. (Correction - measureably more ambivalent and less reassured than before ny visit)
 
Antony Newman said:
Blair,

One question I asked myself was - why would Apple pump so much money into making the hardware for an Apple Watch, when they only sell a few million of them? It would not surprise me if the HW was made at a loss - BUT, the big bucks are made from their cut from Software sales.

With the MacPro (2013) they went to crazy lengths to design the robots, the extrusion systems, assembly lines, automation - again to build a few million machines. As Mac App purchases can happen outside the store - there is far less recurrent income generated, and viability is highly depend on CPU costs.

With the automated nMP build system, and distribution network already existent, and probably >90% of people with a nMP(2013) not desperately in need of an upgrade - perhaps the question at Fruitcamp is firstly, what to they need for people to trigger an wave of upgrades from existing owners? And secondly, what would Apple want to exist in a new workstation to compliment their future plans (before they would be willing to roll-out a replacement).

AJ
Click to expand...

Antony, I would not be surprised if Apple did the tootlimg to save money on.manufacturing Mac Pro's and are making profit to pay for it. Millions, is a pretty big order discount in the server space.

The watch prestige, plus, much of the watch technology has been working out for decades. So all they have to do is put a big price on something with few expensive to develop parts. I have a $17 smart watch phone that may have main parts with only a few million sales. So, a few million at just $100 net, is a few hundred million. At a thousand dollers net, a few billions. Even at a tenth, that's a foot in the door.
 
You must log in or register to reply here.
Back
Top