Since Mac Pro 2013 is officially a failed product...

[flat five]

Explain how opposite was it. Can't? Carry on.

like i said earlier, i don't even know what your point is.

is your point "thunderbolt adoption isn't as widespread as intel initially hoped so they decided to offer royalty free licensing as well as built in controllers" ?

that's what i gather from post #95.

---
if that's what your point is, are you just sharing info (the same info that was shared in #67, 69, & 86)?
or are you saying this in opposition to what others are saying? (like, you're arguing with this info?)
or what?
ie- what is your point?

 

[mavericks7913]

What computers are they using?

Even in school with full of Mac Pros, they dont use TB related prdouct

like i said earlier, i don't even know what your point is.

is your point "thunderbolt adoption isn't as widespread as intel initially hoped so they decided to offer royalty free licensing as well as built in controllers" ?

that's what i gather from post #95.

---
if that's what your point is, are you just sharing info (the same info that was shared in #67, 69, & 86)?
or are you saying this in opposition to what others are saying? (like, you're arguing with this info?)
or what?
ie- what is your point?

Telling the fact: TB is not widely spread yet. Mac Pro 2013 with TB2 was not successful because only a few places used it. and it was limited to pricey products. Even now, TB3 is not a common port and that's why Intel wanted to make royalty free and remove TB controller and then adapt it to CPU.

 

[antonis]

Has nothing to do with Apple. When Thunderbolt is wide spread enough, you will need an Intel processor to utilize it. Intel did it for their own sake, not Apples. Apple was just a side effect.

Agreed, of course. Intel did this for their own shake, in an effort to make TB more mainstream. But given the fact that apple has already invested in TB more than anyone else in their machines (so far), it's them that they should be happy for Intel's decision, more than any other vendor.

 

[linuxcooldude]

Agreed, of course. Intel did this for their own shake, in an effort to make TB more mainstream. But given the fact that apple has already invested in TB more than anyone else in their machines (so far), it's them that they should be happy for Intel's decision, more than any other vendor.

While I'm not totally disagreeing with you, after much thought, how exactly will mass adoption on the PC help Apple? Sure, it will save them some money on manufacturing costs and licensing. Will it make professional TB accessories cheaper? Doubt it. If you want fast speeds your going to have to pay for it. Besides we have USB 3.0\3.1 for that with cheaper prices. Will it make more accessories? Well we already have pro TB devices over a hundred by now, if not more. Yes, one port for slower devices, but will have adapters for that. Majority of consumers on PC side won't make much difference besides smaller ports that can run everything. In fact I think it will have an opposite effect on Apple by making more competition against Apple. Cheap desktops/laptops with TB for budget workstations in the audio/video market, besides the more expensive TB access speeds.

 

[deconstruct60]

...
Telling the fact: TB is not widely spread yet. Mac Pro 2013 with TB2 was not successful because only a few places used it. and it was limited to pricey products. Even now, TB3 is not a common port and that's why Intel wanted to make royalty free and remove TB controller and then adapt it to CPU.

Telling of almost nothing.

1. It isn't going to be integrated across the CPU line up. In fact would be surprising if this isn't solely into PCH chip. It is extremely unlikely that this would be implemented on a x86 core die that doesn't also have the PCH functionality also. it isn't "CPU" where this intergation is being targeted although that would include overall CPU packages where the package isn't primarily just a CPU. Intel does package the PCH chip into a MCM models at the smaller version. However, this is primarily driven by overall market pressure from devices in that category. For example, Qualcomm ragging on size of implemention of Intel's solution.

http://www.anandtech.com/show/11477/qualcomm-announces-first-oems-for-windows-10-on-snapdragon-835

Another case in point the MacBook's almost credit card sized motherboard. That's where TB integration is going to likely play out on the computer system side.

The problem with putting TB in the CPU package is that the CPU package will have to move closer to the port. If it is a mostly a contest of how small you can make the main logic board, that is not going to be a huge problem. For high TPD CPU package that are not going pulled close to the edge of the system that isn't going to work.


2. Similar forces different reasons on the peripheral side. TB+USB+SATA chips , TB+2-5Gb Ethernet chips , TB+USB+Audio+Ethernet+DisplayPort chip (i.e., all the ports for simple dock on a single chip), etc. I suspect the 3rd party implementations to show up in peripherals before showing up on the "host computer" side for Intel based systems.


Sorry both of those are highly indicative that are out of the initial volume ramp stage and into the stage large enough so that the individual niches inside the space can have their own specialty chips to drive the costs down more. ( more almost 85+% solutions out of the box with the provided chip. ). In other words, Intel will have both discrete TB controllers and a few integrated one. So they are transitioning from one "do everything" product to specialized one. You don't do that for a 'too small' market.


Pragmatically TB has subsumed USB since relying on the Type-C port. That means USB-IF practices like multiple implementors pragmatically means coming along too. Intel is taking it slow opening up the implementor/certification process, but given the somewhat quirky roll out of TBv3 docks that really seems to be a prudent idea.

The open specs should like AMD get their own 'boot firmware' done while they work on getting a computer system chip solution done over an extended period of time. ( Unless someone is throwing giant sacks of money at them. I doubt they are in a rush to integrate TBv3 into their CPU packages. )


3. At this point TBv3 (Type-C) is about at least as common in the price range that Apple operates in as mini-DisplayPort was before TBv1-2 picked it up. Picking USB Type-A as the baseline for "common" is highly dubious. Microsoft's Surface Laptop is odd right now because it doesn't have TBv3 not because it is USB Type-A 3.0 only.
(last week's Computex. ASUS VivoBook TBv3 no. ASUS ZenBook Pro & Deluxe TBv3 yes. )


Is Thunderbolt v3 going to roll out in a wave over the $200-700 system market? Probably not. That can't possibly be the only metric for success though.

 

[mavericks7913]

Telling of almost nothing.

1. It isn't going to be integrated across the CPU line up. In fact would be surprising if this isn't solely into PCH chip. It is extremely unlikely that this would be implemented on a x86 core die that doesn't also have the PCH functionality also. it isn't "CPU" where this intergation is being targeted although that would include overall CPU packages where the package isn't primarily just a CPU. Intel does package the PCH chip into a MCM models at the smaller version. However, this is primarily driven by overall market pressure from devices in that category. For example, Qualcomm ragging on size of implemention of Intel's solution.

http://www.anandtech.com/show/11477/qualcomm-announces-first-oems-for-windows-10-on-snapdragon-835

Another case in point the MacBook's almost credit card sized motherboard. That's where TB integration is going to likely play out on the computer system side.

The problem with putting TB in the CPU package is that the CPU package will have to move closer to the port. If it is a mostly a contest of how small you can make the main logic board, that is not going to be a huge problem. For high TPD CPU package that are not going pulled close to the edge of the system that isn't going to work.


2. Similar forces different reasons on the peripheral side. TB+USB+SATA chips , TB+2-5Gb Ethernet chips , TB+USB+Audio+Ethernet+DisplayPort chip (i.e., all the ports for simple dock on a single chip), etc. I suspect the 3rd party implementations to show up in peripherals before showing up on the "host computer" side for Intel based systems.


Sorry both of those are highly indicative that are out of the initial volume ramp stage and into the stage large enough so that the individual niches inside the space can have their own specialty chips to drive the costs down more. ( more almost 85+% solutions out of the box with the provided chip. ). In other words, Intel will have both discrete TB controllers and a few integrated one. So they are transitioning from one "do everything" product to specialized one. You don't do that for a 'too small' market.


Pragmatically TB has subsumed USB since relying on the Type-C port. That means USB-IF practices like multiple implementors pragmatically means coming along too. Intel is taking it slow opening up the implementor/certification process, but given the somewhat quirky roll out of TBv3 docks that really seems to be a prudent idea.

The open specs should like AMD get their own 'boot firmware' done while they work on getting a computer system chip solution done over an extended period of time. ( Unless someone is throwing giant sacks of money at them. I doubt they are in a rush to integrate TBv3 into their CPU packages. )


3. At this point TBv3 (Type-C) is about at least as common in the price range that Apple operates in as mini-DisplayPort was before TBv1-2 picked it up. Picking USB Type-A as the baseline for "common" is highly dubious. Microsoft's Surface Laptop is odd right now because it doesn't have TBv3 not because it is USB Type-A 3.0 only.
(last week's Computex. ASUS VivoBook TBv3 no. ASUS ZenBook Pro & Deluxe TBv3 yes. )


Is Thunderbolt v3 going to roll out in a wave over the $200-700 system market? Probably not. That can't possibly be the only metric for success though.

https://www.wired.com/2017/05/intels-plan-thunderbolt-3-things/

All of your infos are false.

1. They will be going to put TB controller in CPU
2. Read the link.
3. Not even close to 'Common' base on Intel's research

 

[AidenShaw]

While I'm not totally disagreeing with you, after much thought, how exactly will mass adoption on the PC help Apple? Sure, it will save them some money on manufacturing costs and licensing. Will it make professional TB accessories cheaper? Doubt it. If you want fast speeds your going to have to pay for it. Besides we have USB 3.0\3.1 for that with cheaper prices. Will it make more accessories? Well we already have pro TB devices over a hundred by now, if not more. Yes, one port for slower devices, but will have adapters for that. Majority of consumers on PC side won't make much difference besides smaller ports that can run everything. In fact I think it will have an opposite effect on Apple by making more competition against Apple. Cheap desktops/laptops with TB for budget workstations in the audio/video market, besides the more expensive TB access speeds.

You've made some good points.

Just like USB 2.0 killed 1394 by being "good enough" for external strorage, USB 3.0/3.1 will be "good enough" for SSD-based external storage. Only the niche market of higher-end external RAID controller-based storage will need TB-3.

The eGPU market will be a niche of a niche. Windows systems with PCIe slots will have no use for eGPUs. Windows laptop users with CUDA requirements might go with an eGPU - or them might go with an SFF system with a PCIe slot.

Intel probably looks at T-Bolt as a failed bet on technology - so might as well make it free and open source.

 

[linuxcooldude]

Intel probably looks at T-Bolt as a failed bet on technology - so might as well make it free and open source.

I'm still wondering why people think Thunderbolt for pro use has to have a wide adoption rate to succeed. Wide adoption works when you need a multitude of cheap devices to work on a variety of platforms like USB. You don't need wide adoption when a higher cost to those who need the fastest speeds and unique speciality for those who are willing to pay for it.

I think Intels looking at Thunderbolt USB-C in its CPU's and Thunderbolt for professional use differently. While TB3 in USB-C can be useful in consumer PC's by the average user as it covers so many standard protocols cheaply in a small miniaturized adapter which is handy for smaller devices. Obviously the tie in with a high adoption rate is necessary for it to work in consumer PC's and will require an Intel CPU to make it happen so Intels making money either way.

I don't think Intel really sees Thunderbolt as a failed bet, but how to make more money off of it in the consumer realm when most consumers are not willing to spend that much money. I think they found that way through USB-C. But in the end the consumers win with this type of adoption.

 

[deconstruct60]

https://www.wired.com/2017/05/intels-plan-thunderbolt-3-things/

All of your infos are false.

There is a difference between you wishing it false and that actually being the case.

1. They will be going to put TB controller in CPU

I never said there would be zero "CPU" packages with integrated TB. Primary point is inclusion, or not, of the PCH (I/O Hub) is likley going to be indicative of where it is incorporated or not. There are no points where there is a quote from Intel stating that all of their CPU products are getting it. There are references to "CPUs" but that actually is not a universal qualification no matter how hard you flap your arms.

2. Read the link.

So regurgitation from the tech porn press is 'proof'? Chuckle. How about we actually use direct Intel quotes with substantive context? Here Intel's press release/blog about the topic.

https://newsroom.intel.com/editorials/envision-world-thunderbolt-3-everywhere/

"... it plans to drive large-scale mainstream adoption of Thunderbolt by integrating Thunderbolt 3 into future Intel CPUs and by releasing the Thunderbolt protocol specification to the industry next year. ..."

Not all Intel CPUs but "future CPUs". The adjective there is 'future' not an enumeration or qualification of the whole Intel CPU product line. So what is the rest of their story.......


a. Why integrate .....

"... With Thunderbolt 3 integrated into the CPU, computer makers can build thinner and lighter systems with only Thunderbolt 3 ports. ... "

What? You mean like the one port wonder the MacBook? Or the 1-2 port wonders like those Intel and Qualcomm boards I linked in about. Intel you mean like those? Thinner and lighter means desktops ,workstation, and desktop replacement laptops ? Probably not. So false? ..... not even close.

3. Not even close to 'Common' base on Intel's research

the quote

".... Well on our way to achieving our vision, ..." No one said they were done. But there is no statement by Intel that they are stuck in the same place there were when TB launched. It is a drive up to the next level..... which it itself probably isn't at the mundane mainstream.


c. So what kind of applications is Intel thinking this port brings.

Common like......
" State-of-the-Art Single-Cable Docks ".

So if went from $299+ docks to $199 docks (with specialized dock implementation chips) how many folks buying $499-599 systems are going to buy $199 docks? That is going to be common? Really? Compare that to the market were $1,000K systems have $199 docks... how common is that going to be. ( in other words what is likelihood that folks who buy $1000+ systems have an additional $199 in disposable income? ). So false? .... not even close.

Most desktops like the Mac Pro need a dock because ??? Not really. Is it highly useful and effective for the Mac Pro tap into the same ecosystem as the rest of the Mac product line up? Absolutely. The "light and low power" integration needs though are no where near as high.


"Faster-than-Ever Storage "

Same issue. Bleed edge NVMe storage. Buying significantly large capacities of that to be used in sneak net movement of data. The $599 crowd or the $1000 crowd.

NVMe is just PCI-e. if make a specialized one port controller just to hook to a drive how is that going to be more expensive than a PCI-e to USB 3.1 gen 2 converter. Plus not stuck with USB latencies.
Two ports, yeah the controller will be more expensive but price point on Apple's TB Ethernet and FW adaptors was not that high. ( P.S. for folks claiming having never seen TB in action .... I highly doubt they counted the those two adapter usage. If looking for big disks subsystems and expensive docks. Perhaps. But any MBP of last couple years hooked to Ethernet was??? )


" Epic Gaming ...enjoy long hours at a coffee shop with a thin and light notebook. Then, when it’s time for epic gaming, the notebook can be paired with a Thunderbolt 3 external graphics box "

Well there is that pesky highly thin and light optimized system again. Oh no that can't possibly be what this next growth objective is focused on. False , false , false ........ not!

Maybe they shave $100 of the external GPU enclosure prices .... that is still going to leave them in the $199-299 range. Yet again for $499-599 systems what is the market for another $199 before even buy the "epic gaming" GPU card????

" Lifelike Virtual Reality .... Only Thunderbolt 3 can deliver up to 4K resolution in virtual reality with a single cable. "

4k to each eyeball. yeah that is going to be inexpensive and lowest common denominator 'common' in the near future... not.



Finally ( which actually is at the beginning of the Intel note on the topic) they seem to be drifting back into the "one port to rule them all" hype machine. I thought it was wrong when they did it for TBv1 and it is essentially just as wrong now.

"... . A world where one USB-C connector does it all – today, and for many years to come. ...'

The notions that the overall broad set of systems need one and only one port type. ( or one and only one port as in iOs devices and MacBooks ) has major problems. Trying to push the Mac desktops into a corner where they do not have Ethernet ports is just a plain bozo move. Even more so for the power cord for largely fixed in place systems. The notion that USBType-A is going to rapidly fade away in the next 1-2 years .... borders on delusional (most TB docks are just going to displace Type-A ports in space .... not make they completely disappear. There isn't a single dock that doesn't have Type-A and that probably isn't going to change over next 2-5 years. )

Thunderbolt needs a "peep rally" press release. But some of the hype here is just obvious if not just resigned to letting Intel spin blow tons of smoke (which the vast majority of the tech porn press articles did. )


P.S. As far as the side effect of the TB controller would have if coupled to the CPU package... This is from TBv1-2 but extremely likely still valid with TBv3.

".... The spec for max trace length between the Thunderbolt controller and port is two inches, compared to up to 10 inches for Intel's USB 3.0 controller. "
http://www.anandtech.com/show/5884/...s-part-2-intels-dz77rek75-asus-p8z77v-premium

Go and look at MBP , Mac Pro , Mini and iMac tear downs. Is the controller ever more than two inches away from the port. Go look at other boards, add in cards, etc. See any violations?

If look at higher end ATX boards where multiple 6Gbps SATA and/or NVMe PCI-e M.2 sockets. If the PCH controller closer to those or to the mundane USB 2.0 sockets on the back edge plate?

Folding a PCH with very high bandwidth issues into a CPU package is going to place restrictions on that package placement. There is not Intel article or tech porn press article that stipulates that the placement constraints have been lifted. They haven't.

Which if trying to deal with 60-180W CPUs you'd be dragging them within 2 inches of the edge of the PC.... that is not a good place for that kind of thermal. It just isn't. Which is why it likely wouldn't be integrated there.

Even without 60-180W CPUs if you want TBv3 ports on either side of your 11 inch wide laptop and the CPU+TB controller has to be within 2 inches of either side ..... there is inotihing in these press blitz that solves that problem either. Discrete TB controllers are going to be siting around for a long time because integration if the opposite of a solution in that context.

 

[antonis]

While I'm not totally disagreeing with you, after much thought, how exactly will mass adoption on the PC help Apple? Sure, it will save them some money on manufacturing costs and licensing. Will it make professional TB accessories cheaper? Doubt it. If you want fast speeds your going to have to pay for it. Besides we have USB 3.0\3.1 for that with cheaper prices. Will it make more accessories? Well we already have pro TB devices over a hundred by now, if not more. Yes, one port for slower devices, but will have adapters for that. Majority of consumers on PC side won't make much difference besides smaller ports that can run everything. In fact I think it will have an opposite effect on Apple by making more competition against Apple. Cheap desktops/laptops with TB for budget workstations in the audio/video market, besides the more expensive TB access speeds.

The market is not so forgiving, though. Up until now TB wasn't a successful standard, it needs a higher adoption rate. Apple has put their eggs in the TB basket by the moment they resigned from internal upgrades path. PC ecosystem doesn't care as much, since they still have the internal upgrades, apple does not. PCs are nearly ignoring TB up until now, and this won't make the TB standard easier to expand. Yes, I could bet that prices will drop, since it's the PCs that always drive the mass market, if more and more manufacturers start to produce TB products (which until now are not that much available in the rest of the world). Competition is good in that matter. TB needs the market share and a higher adoption rate or it will be dead. It needs to be more common (whatever this includes; more and cheaper products, from more manufacturers etc).

 

[linuxcooldude]

Up until now TB wasn't a successful standard, it needs a higher adoption rate. Apple has put their eggs in the TB basket by the moment they resigned from internal upgrades path.

Your talking generally. Tell me specifically why does Thunderbolt need mass adoption to be successful in the professional market? Not everyone needs the fastest speeds or is a professional needing Thunderbolt. The higher costs justify the need if your in that market, not high adoption rates.

PCs are nearly ignoring TB up until now, and this won't make the TB standard easier to expand.

The majority of PC's don't cater to the pro market, hence not the need for a wide adoption rate in PC's. Average consumers won't pay for a professional periphrials if they don't need it or can get it cheaper using USB 3.0 if its good enough.

The reason why Thunderbolt will work using USB-C in PC's is it still provides most other standards at a lower cost: USB 2.0/3.0/3.1/HDMI/DisplayPort/DVI ect. Most PC users won't be buying High end Thunderbolt devices, but will still be able to use Thunderbolt USB-C for everything else. In that case mass adoption rate is necessary. But high adoption rate for Thunderbolt for use in a niche pro market won't be needed.

 

[lowendlinux]

It's failed and now you're getting an iMac Pro

 

[antonis]

Your talking generally. Tell me specifically why does Thunderbolt need mass adoption to be successful in the professional market? Not everyone needs the fastest speeds or is a professional needing Thunderbolt. The higher costs justify the need if your in that market, not high adoption rates.



The majority of PC's don't cater to the pro market, hence not the need for a wide adoption rate in PC's. Average consumers won't pay for a professional periphrials if they don't need it or can get it cheaper using USB 3.0 if its good enough.

The reason why Thunderbolt will work using USB-C in PC's is it still provides most other standards at a lower cost: USB 2.0/3.0/3.1/HDMI/DisplayPort/DVI ect. Most PC users won't be buying High end Thunderbolt devices, but will still be able to use Thunderbolt USB-C for everything else. In that case mass adoption rate is necessary. But high adoption rate for Thunderbolt for use in a niche pro market won't be needed.

Professional market is just a term. In reality there are various levels of needs, and the line between a standard user, a semi-pro user, a hobbyist and a professional is very blurry. Having more options and making something more common (instead of aiming only in a niche market) will bring a bigger variety in prices as well, allowing every one of the aforementioned categories to have access to this technology. It opens the technology up to a wider range of users, creates competition and options.

Or, in other words, Intel had a vision for TB. They wanted it to be the one port of everything and - technologically wise - they succeeded. But the current state of TB hardly fulfills their vision. It is obvious they want this to be adopted in a much wider range by opening the standard.

From the article linked above:

Making Thunderbolt 3 easier to incorporate may help drive adoption, but lowering the cost would as well.

and:

“Cost is always a consideration,” says Ziller. “I think the integration into future CPUs will help reduce the overall solution cost on the computer. And we’re continually working with the industry to lower the cost of the cables and the devices.” He notes, too, that improving USB-C economies of scale should help Thunderbolt 3 drive down costs as well.

This will definitely help a vast number of users (including mac users) to reach this technology, hence macs will be more relevant to a higher number of people.

 

[ActionableMango]

However, I refute a poorely defined conclusion based on no evidence. I know what Apple management team said at the meeting but they did not back it up with hard evidence as far as I know, meaning that it is impossible for an independent source (the tech press or you) to verify that Apples management team has drawn the "correct" conclusions.

I suppose that is true, but requiring evidence that only an insider could possibly have is an unreasonably high bar to reach for an Internet discussion forum on a rumors website.

I will point out that the few people who do have access to that sort of information have decided to cancel the current design. That's good enough for most to conclude that it is a failure. If your standards of evidence require detailed insider information to be revealed before you can come to a conclusion, then nearly all discussion in these forums must be unsatisfactory to you.

 

[macsforever]

Im expecting new Mac Pro to be better but I really doubt about if they gonna make upgradable Mac Pro since all Mac computers are all in one computer which is no upgradable for a while. Even they do make one, MacOS need to support softwares too. I really wish to see liquid cooler in Mac computer since it performed way better than air cooler which I tried before. We know that PC can be built with less amount of money but perform much better than Mac Pro or others. I really think that Apple should do something about mac pro. If it's just too expansive with ridiculous design like Mac Pro 2013, they are dumb.

Yeah right!
The new imac pro has already smashed price and performance issues... PC's are done!

 

[mavericks7913]

Yeah right!
The new imac pro has already smashed price and performance issues... PC's are done!

Well although iMac Pro is impressive but still limited. You can not upgrade and expand at all. iMac Pro is all I need but for other users who need high performance, it wont be enough. I really doubt about its cooling system already.

 

[AidenShaw]

Yeah right!
The new imac pro has already smashed price and performance issues... PC's are done!

You must have set your time machine to January 2018 - because that's probably when volume shipments will start.

 

[ZombiePhysicist]

Yeah right!
The new imac pro has already smashed price and performance issues... PC's are done!

I don't think you're being nearly positive and optimistic enough. You're like maybe at a screaming 6 on the 'reality has nothing to do with my optimism scale.' The scale goes to 11. 11.

 

[linuxcooldude]

I will point out that the few people who do have access to that sort of information have decided to cancel the current design. That's good enough for most to conclude that it is a failure. If your standards of evidence require detailed insider information to be revealed before you can come to a conclusion, then nearly all discussion in these forums must be unsatisfactory to you.

Well I would say 95% of the things posted are speculation. So far it seems in the past most people considered the Mac Pro as dead, and now Thunderbolt too. We all know now it isn't true.

 

[AidenShaw]

Well I would say 95% of the things posted are speculation. So far it seems in the past most people considered the Mac Pro as dead, and now Thunderbolt too. We all know now it isn't true.

To be fair, T-Bolt as in "T-Bolt" and "T-Bolt 2" is nearly comatose.

T-Bolt 3 on USB-C and built into the PCH or CPU is a potentially a completely different beast. Intel heard the death rattle from T-Bolt and essentially started over.

 

[mrxak]

It is way too early to judge the iMac Pro. We don't know how loud it will be. We don't know what kind of thermal throttling it might require. Until I see real world benchmarks, I'm not going to make any sort of guess how popular it will be or how good a stop-gap it will be until Apple releases the new new Mac Pro.

 

[AidenShaw]

It is way too early to judge the iMac Pro. We don't know how loud it will be. We don't know what kind of thermal throttling it might require. Until I see real world benchmarks, I'm not going to make any sort of guess how popular it will be or how good a stop-gap it will be until Apple releases the new new Mac Pro.

My worry would be that the iMac Xeon has a 500 watt power supply. That's really tiny for a workstation, especially one with an 18 core Xeon and a "workstation class GPU". The MP6,1, more than anything else, was killed by its 450 watt power budget.

Even if the thermals are perfect so that there's no thermal throttling - everything may be underclocked to fit in that small 500 watt envelope, so there's power throttling.

500 watts? That's really sad.

But we won't really know until the very end of the year or early 2018.

 

[saulinpa]

The new imac pro has already smashed price and performance issues... PC's are done!

That is another dream as the iMac Pro smashed price the wrong way. Base model starting at $5000. That's 66% more than the intro price of the nMP.

Apple is not going after the pro market. They proved that they don't care about pros. They are simply going after those with money.

 

[linuxcooldude]

To be fair, T-Bolt as in "T-Bolt" and "T-Bolt 2" is nearly comatose.

Advancements in current standards don't mean it dead, just improved. As dually noted by another poster, its just semantics.

T-Bolt 3 on USB-C and built into the PCH or CPU is a potentially a completely different beast. Intel heard the death rattle from T-Bolt and essentially started over.

Even without USB-C and integration into Intels CPU, Thunderbolt 3 is already different. It uses fewer lanes to get faster speeds. USB-C & integration just changes its implementation, it doesn't change so much as a protocol.

Intel heard the death rattle from T-Bolt and essentially started over.

If Thunderbolt is so expensive, not widely used, a so called dead standard, why would Intel feel they need to save it?!? Why risk losing licensing and royalty fees if its not making a profit some where? Because its very important to intel and the high end market runs on it in some form. Now its going to run on a multitude of low and middle end PC products.

Even if Thunderbolt went away entirely, what would replace it that is just as fast and flexible as Thunderbolt?

Or, in other words, Intel had a vision for TB. They wanted it to be the one port of everything and - technologically wise - they succeeded. But the current state of TB hardly fulfills their vision. It is obvious they want this to be adopted in a much wider range by opening the standard.

Who's to say USB-C was Intels way of implementing Thunderbolt from the very beginning. The USB-C standard was created by the USB-IF forum. And what two formidable members are part of that forum?!? Thats right, Intel & Apple. You can't go straight from Lightpeak to Thunderbolt to USB-C overnight. it takes lots of time, money and research.

That is another dream as the iMac Pro smashed price the wrong way. Base model starting at $5000. That's 66% more than the intro price of the nMP.

The nMP does not come with a 27" 5K Retina display. You have to add the display as part of the base iMac price. A lot of people seem to forget that.

 

[saulinpa]

The nMP does not come with a 27" 5K Retina display. You have to add the display as part of the base iMac price. A lot of people seem to forget that.

They forget that because a lot of people don't want the display. So whether you want one or not you have to pay for the display. Just more Apple not knowing what people want.