Jump to content

US$189 AKiTiO Thunder2 PCIe Box (16Gbps-TB2)


Recommended Posts

That is great news gualque, thank you very much for your testing :)

It looks like we've got a x4 PCIe connector there, but just to be sure, let's wait for Tech Inferno Fan's conclusion :)

From the results so far it does look like it the bottom connector is x4.

How to easily supply 75W slot power, bypassing the need for a PCIe riser? Done by solder a female molex (12V/GND) directly to the 12V pins on x4 slot outer edge. That would be near the circle you've marked below. If no GND pin is nearby, the just tap any of the four screwhole posts or the screws themselves. All of them are attached to the GND plane. By the looks of it, those two solder points could be accessed without even needing to dismantle the whole enclosure. If using a screw then you've only soldered to one point on the systemboard. That being 12V in.

[OPTIONAL]: How to eliminate the use of the 60W AC adapter?

Optional because the simplest way of powering the TB board is to use the supplied 60W AC adapter,

Can you also advise if a standard 4-pin floppy molex lead from an ATX PSU (WARNING: must be switched OFF!!!) plugs into that connector on the Thunderbolt board where the yellow/black cable attaches, or could do with some very simple mods? If so, then note the floppy molex has a yellow-black-black-red pinout. Does at least one yellow and one black match the yellow/black cable pinout positions when placed onto of each other for comparison, orientated the same way?

I ask as neutering a floppy molex lead from an ATX PSU with just the single matching yellow & black could provide the direct power to feed the TB board. That would be a cheaper workaround than getting a molex-to-DC jack to provide that 12V via the DC jack, then across the yellow/black cable to the TB board.

Now very promising news! B1 connector (rear side J10) on the right corner of "x4":

[ATTACH=CONFIG]12324[/ATTACH]

8 small holes through the board, 7 of them gave continuous beep sound with PCIe 2nd pin [12V], one is different but couldn't get GND continuity to it. The nearest GND is marked on the picture, that point is very small, but I was able to get continuous beep when paired with a screw hole.

Link to comment
Share on other sites

Building on Arclite's idea of a female-female PCIe 4x to x16 powered riser, can we connect 2 of the following PCIe x16 powered risers together using only the female portions? The cables are detachable. We will have a female-female PCIe x16 to x16 powered riser - one for the TB board (if it is a functional PCIe 4x card) and the other for the graphics card.

post-27765-14494998204729_thumb.jpg

  • Thumbs Up 1
Link to comment
Share on other sites

Excellent news on the x4 connector. I would be really great if we could replace the PCIe adapter board entirely, both for warranty concerns and also for powering hungrier GPUs without issues. The concern I have with hooking up a molex to feed more power to the board next to the TB connection is that the tracks might not be spec'd for carrying that much current. My experience with a fried PE4H v1 because of a GTX 470 trying drawing more than 75W makes me a bit wary of trying this. Someone here will probably have a more informed opinion on whether the tracks on this board will be up to spec.

With a standard x4 connector very likely in place, the molex connector between boards is the final puzzle to work out a complete PCIe riser replacement. Besides just supplying power it could also be transmitting data, similarly to the "special" header on the ThunderboltEX II card. Can we tell the difference with a multimeter or do we need a different instrument (e.g. oscilloscope)?

Building on Arclite's idea of a female-female PCIe 4x to x16 powered riser, can we connect 2 of the following PCIe x16 powered risers together using only the female portions? The cables are detachable. We will have a female-female PCIe x16 to x16 powered riser - one for the TB board (if it is a functional PCIe 4x card) and the other for the graphics card.

[ATTACH=CONFIG]12329[/ATTACH]

That's a nice idea, although it would be great if at least one of the female connectors was x4 so it wouldn't protude out of the top of the enclosure when connected to the TB board. I'm considering soldering a female half on a x4 to x4 riser cable but it's going to be tedious work.

Link to comment
Share on other sites

Can you also advise if a standard 4-pin floppy molex lead from an ATX PSU (WARNING: must be switched OFF!!!) plugs into that connector on the Thunderbolt board where the yellow/black cable attaches, or could do with some very simple mods? If so, then note the floppy molex has a yellow-black-black-red pinout. Does at least one yellow and one black match the yellow/black cable pinout positions when placed onto of each other for comparison, orientated the same way?

I ask as neutering a floppy molex lead from an ATX PSU with just the single matching yellow & black could provide the direct power to feed the TB board. That would be a cheaper workaround than getting a molex-to-DC jack to provide that 12V via the DC jack, then across the yellow/black cable to the TB board.

I already have a self made molex to 4-pin yellow/black cable to the TB card, and I have tested it. Look at here. A standard 4-pin floppy molex plug was a little too big (but the holes seemed to have correct positions with the TB card's 4 pins). Those test results make me think that we need some kind of "loop" connection via the 4-pin yellow/black cable to the x16 PCIe slot. I still don't know what is the meaning of a very short beep sound when I did that continuity test from near DC jack points to the x16 PCIe 2nd pin. So before bypassing the PCIe board entirely, feeding 75W from the x4 connector to the x16 slot may be more workable solution.

EDIT: For those who are interested about female-to-female-board/riser-solution, I found these:

http://www.boston.co.uk/products/components/c.$1655$7120/default.aspx

PCIe x8 1.0 Expansion Backplane (184) | MaxExpansion.com

RSC-R2UT-2E8R looks quite interesting. I asked from them, and it has a 6-pin power connector (75W). This is just an idea... but would they help in theory? If not, maybe they can be used for SLI mode or for fast PCIe SSD :)

Link to comment
Share on other sites

I already have a self made molex to 4-pin yellow/black cable to the TB card, and I have tested it. Look at here. A standard 4-pin floppy molex plug was a little too big (but the holes seemed to have correct positions with the TB card's 4 pins). Those test results make me think that we need some kind of "loop" connection via the 4-pin yellow/black cable to the x16 PCIe slot. I still don't know what is the meaning of a very short beep sound when I did that continuity test from near DC jack points to the x16 PCIe 2nd pin. So before bypassing the PCIe board entirely, feeding 75W from the x4 connector to the x16 slot may be more workable solution.

EDIT: For those who are interested about female-to-female-board/riser-solution, I found these:

http://www.boston.co.uk/products/components/c.$1655$7120/default.aspx

PCIe x8 1.0 Expansion Backplane (184) | MaxExpansion.com

RSC-R2UT-2E8R looks quite interesting. I asked from them, and it has a 6-pin power connector (75W). This is just an idea... but would they help in theory? If not, maybe they can be used for SLI mode or for fast PCIe SSD :)

The idea proposed earlier does not bypass the 4 pin yellow/black cable connector on the TB board.

Instead what I was doing was suggesting was:

1. supplying the ~35W (60W-25W=35W remaining) to the TB board either via the stock 60W AC adapter, custom molex-to-barrel or a modified floppy molex cable.

2. 75W would be provided to the PCIe slot directly bypassing the x4 connector. That bypass circumventing any power restricting fuses the TB board might have as well as avoiding passing up to 75W across the x4 connector that's probably not specced to carry that much. AKiTiO speccing it at 25W. The bypass done either using a powered PCIe rise plugged into the x16 slot, soldering 12V/GND directly on the AKiTiO PCIe board or coming up with an alternate powered x4-to-x16 female ended board to replace the AKiTiO PCIe one.

Link to comment
Share on other sites

The idea proposed earlier does not bypass the 4 pin yellow/black cable connector on the TB board.

Instead what I was doing was suggesting was:

1. supplying the ~35W (60W-25W=35W remaining) to the TB board either via the stock 60W AC adapter, custom molex-to-barrel or a modified floppy molex cable.

2. 75W would be provided to the PCIe slot directly bypassing the x4 connector. That bypass circumventing any power restricting fuses the TB board might have as well as avoiding passing up to 75W across the x4 connector that's probably not specced to carry that much. AKiTiO speccing it at 25W. The bypass done either using a powered PCIe rise plugged into the x16 slot, soldering 12V/GND directly on the AKiTiO PCIe board or coming up with an alternate powered x4-to-x16 female ended board to replace the AKiTiO PCIe one.

Ok, thanks! I was actually referring what ithildin said: "I would be really great if we could replace the PCIe adapter board entirely, both for warranty concerns and also for powering hungrier GPUs without issues."

Link to comment
Share on other sites

Now very promising news! B1 connector (rear side J10) on the right corner of "x4":

[ATTACH=CONFIG]12324[/ATTACH]

8 small holes through the board, 7 of them gave continuous beep sound with PCIe 2nd pin, one is different but couldn't get GND continuity to it. The nearest GND is marked on the picture, that point is very small, but I was able to get continuous beep when paired with a screw hole.

EDIT: Voltage metering also proved that this is the right place to feed extra power. All those 7 points gave +12V. Is this a standard size power connector?

The more I look at the TB board, the more it resembles a PCIe x4 TB card like the one from MSI.

https://thunderbolttechnology.net/product/msi-thunderbolt-plus

Link to comment
Share on other sites

I am interested as well

News?

I send the Akitio Box back to Macway, neither my R9 290 nor my old GTX 660 ti worked properly. I suppose that my Akitio Thunder 2 was broken, because most of the time I got a black screen on my external display (no source avaible for both graphic cards, although my device manager listed the unplugged devices after restart). Also optimus did'nt seem to be working for my old GTX 660 ti the few times I got a stable connection.

Link to comment
Share on other sites

I double checked the x4 slot and was so sure that made a real test by a x4-male to x16-female molex-powered raiser, x16 end directly attached to the TB card, and x4 to the PCIe board. It worked perfectly with my GTX 780 and passed 3DMark11 test with similar results as earlier. One important note: the GPU has to be powered via the raiser only. I do not recommend this as a solution, because it may be unsafe.

As far as I know, x4 slot is specified to feed max 25W (http://en.wikipedia.org/wiki/PCI_Express). Even though PSU 12V rail is able to give more, it could be unsafe. I don't want melted wires. We need 6-pin power plug to feed 75W to the x16 via the PCIe board rails somehow, am I right Tech Inferno Fan? One way is two 4-pin molex plugs to one 6-pin plug adapter as well, and I think I have one. But the question is, how can we provide 75W to the PCIe slot safely? I am a bit worried also about those little holes beside the x4 slot, how much power they can eat safely.

Thanks for all here who have helped me to this point, and hopefully my tests give some great ideas for your future implementations. I myself stay with the solution I have at the moment; no powered risers, no soldering and no AKiTiO's power adapter.

Link to comment
Share on other sites

I double checked the x4 slot and was so sure that made a real test by a x4-male to x16-female molex-powered raiser, x16 end directly attached to the TB card, and x4 to the PCIe board. It worked perfectly with my GTX 780 and passed 3DMark11 test with similar results as earlier. One important note: the GPU has to be powered via the raiser only. I do not recommend this as a solution, because it may be unsafe.

As far as I know, x4 slot is specified to feed max 25W. Even though PSU 12V rail is able to give more, it could be unsafe. I don't want melted wires. We need 6-pin power plug to feed 75W to the x16 via the PCIe board rails somehow, am I right Tech Inferno Fan? One way is two 4-pin molex plugs to one 6-pin plug adapter as well, and I think I have one. But the question is, how can we provide 75W to the PCIe slot safely? I am a bit worried also about those little holes beside the x4 slot, how much power they can eat safely.

Thanks for all here who have helped me to this point, and hopefully my tests give some great ideas for your future implementations. I myself stay with the solution I have at the moment; no powered risers, no soldering and no AKiTiO's power adapter.

In which case you've confirmed the TB board and the connector on the PCIe board are using a x4 layout.

Well, your powered x4-male to x16-female interposer board is another way we can get 75W slot power without needing to solder, or use a riser in the pcie slot itself. Is that a workable solution? Depends how much wiggle room there is to attach that as an interposer (TB<->interposer<->PCIe board) in place of the factory arrangement (TB<->PCIe board).

Certainly, your solution of using an ATX PSU + molex-to-barrel connector to power your GTX780 + slot works for you. That's because it appears to not need 75W slot power and can do with the ~25W supplied to the slot.

That's going to be a hit-and-miss solution. GPUs without pci-e power will not work since they draw slot power only, eg: GTX750. Ones with one or two PCIe power connectors *might* work, depending on how much slot power they draw.

We can see the T|I user dernils, presumably using the same power delivery arrangement, didn't have the same level of success as you did. Neither his R9 290 or GTX660Ti were functional in a AKiTiO Thunder2: http://forum.techinferno.com/diy-e-gpu-projects/7486-cheapest-thunderbolt-egpu-option-europe-2.html#post103783

  • Thumbs Up 1
Link to comment
Share on other sites

In which case you've confirmed the TB board and the connector on the PCIe board are using a x4 layout.

Well, your powered x4-male to x16-female interposer board is another way we can get 75W slot power without needing to solder, or use a riser in the pcie slot itself. Is that a workable solution? Depends how much wiggle room there is to attach that as an interposer (TB<->interposer<->PCIe board) in place of the factory arrangement (TB<->PCIe board).

Certainly, your solution of using an ATX PSU + molex-to-barrel connector to power your GTX780 + slot works for you. That's because it appears to not need 75W slot power and can do with the ~25W supplied to the slot.

That's going to be a hit-and-miss solution. GPUs without pci-e power will not work since they draw slot power only, eg: GTX750. Ones with one or two PCIe power connectors *might* work, depending on how much slot power they draw.

Yes, I was just very lucky with my selected GTX 780 as it doesn't need more than 25W from the x16 slot. Works for me at the moment and I will update my test results and put more detailed instructions in near future. I was very curious to know if there are other solutions to provide that 75W, so that people here may have more luck with other GPUs too :) If there is a workable and safe solution without much soldering, I may try it one day.

The 4-pin yellow/black cable is still compulsory in this test too, and you cannot use the powered riser and power from DC jack (also by using AKiTiO's power adapter) at the same time because in that case, when starting 3DMark11 test, your system will shut down.

Link to comment
Share on other sites

I send the Akitio Box back to Macway, neither my R9 290 nor my old GTX 660 ti worked properly. I suppose that my Akitio Thunder 2 was broken, because most of the time I got a black screen on my external display (no source avaible for both graphic cards, although my device manager listed the unplugged devices after restart). Also optimus did'nt seem to be working for my old GTX 660 ti the few times I got a stable connection.

Your earlier post gave a bit hope that you managed to get R9 290 working?

<quote=dernils;103273>I finally got a used GTX 660 ti for my Akitio Thunder2. Though my external display doesn't recognize a source after a restart and the internal lcd stays black. I had similiar issues with my R9 290, but after a few tries i got it working.

</quote=dernils;103273><quote=dernils;103273>- Can you say what 550W power supply did you use?

- Did you use a "paperclip" method to power up PSU and what kind of x16 to x16 powered riser? There exist risers that don't have black ground wires. I used one with two yellow + two black wires + capacitor.

- Windows 8.1 Boot Camp installation?

I bought mine also from MacWay in France.

</quote=dernils;103273>

Link to comment
Share on other sites

The idea proposed earlier does not bypass the 4 pin yellow/black cable connector on the TB board.

Instead what I was doing was suggesting was:

1. supplying the ~35W (60W-25W=35W remaining) to the TB board either via the stock 60W AC adapter, custom molex-to-barrel or a modified floppy molex cable.

2. 75W would be provided to the PCIe slot directly bypassing the x4 connector. That bypass circumventing any power restricting fuses the TB board might have as well as avoiding passing up to 75W across the x4 connector that's probably not specced to carry that much. AKiTiO speccing it at 25W. The bypass done either using a powered PCIe rise plugged into the x16 slot, soldering 12V/GND directly on the AKiTiO PCIe board or coming up with an alternate powered x4-to-x16 female ended board to replace the AKiTiO PCIe one.

This is the safest solution right?

If we allow the TB board to draw power from the powered PCIe riser plugged into the PCIe x16 slot of the PCIe board, the molex supplying the power could potentially be overloaded with the 75W power required by the graphics card + power required by the TB board? Conversely, if we allow the PCIe board to draw the 75W power from the x4 connector via the TB card, the TB card could potentially be overloaded as well?

- - - Updated - - -

Is your solution running stable?

I'm just wondering, because some people posted, that their (even with a powered riser) solution were unstable and only worked for some time until windows crashes.

I think those were mostly R9 290 and 290X cards.

My Powercolor HD 7870 runs stably but my R9 290X causes Windows to crash after a few minutes.

Link to comment
Share on other sites

This is the safest solution right?

If we allow the TB board to draw power from the powered PCIe riser plugged into the PCIe x16 slot of the PCIe board, the molex supplying the power could potentially be overloaded with the 75W power required by the graphics card + power required by the TB board? Conversely, if we allow the PCIe board to draw the 75W power from the x4 connector via the TB card, the TB card could potentially be overloaded as well?

It will depend on how the powered riser cable was built. The connector itself can withstand about 11A (122W) but a single wire can top out at 5A (60W) which is not sufficient to meet the 75W PCIe spec + power the TB card and may crash or burn out under load. Some powered risers use two 12V wires so in theory they can carry up to 10A (120W) which should be enough for everything.

I think those were mostly R9 290 and 290X cards.

My Powercolor HD 7870 runs stably but my R9 290X causes Windows to crash after a few minutes.

Does your powered riser cable have a solid capacitor soldered on (like this one)? According to bitcoin mining forums, more recent versions of powered risers have capacitors built in to stabilize power supply and prevent system crashes under sustained load.

  • Thumbs Up 1
Link to comment
Share on other sites

It will depend on how the powered riser cable was built. The connector itself can withstand about 11A (122W) but a single wire can top out at 5A (60W) which is not sufficient to meet the 75W PCIe spec + power the TB card and may crash or burn out under load. Some powered risers use two 12V wires so in theory they can carry up to 10A (120W) which should be enough for everything.

Does your powered riser cable have a solid capacitor soldered on (like this one)? According to bitcoin mining forums, more recent versions of powered risers have capacitors built in to stabilize power supply and prevent system crashes under sustained load.

This is the one that I am using. It only has one 12V wire so I guess it supports up to 5A (60W)?

post-27765-14494998206801_thumb.jpg

Link to comment
Share on other sites

Is your solution running stable?

I'm just wondering, because some people posted, that their (even with a powered riser) solution were unstable and only worked for some time until windows crashes.

Very stable. I have done performance tests that I will post here soon, and some occasional gaming. Some games running on 2880 x 1800 native resolution.

Link to comment
Share on other sites

Does this implementation require Windows 8.1? Can it work with Windows 7?

Yes, my implementation requires Windows 8.1 Boot Camp installation. Not sure about Windows 7, because I haven't tested it. May be possible that it doesn't work.

Link to comment
Share on other sites

Yes, my implementation requires Windows 8.1 Boot Camp installation. Not sure about Windows 7, because I haven't tested it. May be possible that it doesn't work.

Thank you. Am I correct in thinking that Windows 8 now boots via EFI when using Boot Camp? I think about a year ago one had to avoid Boot Camp and install manually to enable that.

Link to comment
Share on other sites

Your earlier post gave a bit hope that you managed to get R9 290 working?

<quote=dernils;103273>- Can you say what 550W power supply did you use?

- Did you use a "paperclip" method to power up PSU and what kind of x16 to x16 powered riser? There exist risers that don't have black ground wires. I used one with two yellow + two black wires + capacitor.

- Windows 8.1 Boot Camp installation?

I bought mine also from MacWay in France.

</quote=dernils;103273>

I thought that my R9 290 is working because after about 20 tries I managed to get an unstable (kept crashing) connection to my external display. So I connected my old Zotac GTX 660 ti, especially for optimus. Same as the AMD, after about 20 tries I got a connection, although I never managed to get optimus working. Yea this was my experience with the Akitio Box... never again I got a connection between my Macbook, the gpu (neither AMD nor Nvidia) and the external display, although I tried it like 100 times. This is my setup:

Macbook Retina 13" 2014 (no dgpu)

Powered riser ([PCIE] 16x - 16x PCI-Express POWERED Riser Flex) same riser as yours

a 650 Watt PSU, worked flawless in my desktop computer (http://www.amazon.de/Super-Flower-SF650P-14P-PC-Netzteil-ATX2-31/dp/B002BIRRIM/ref=sr_1_2?ie=UTF8&qid=1409222326&sr=8-2&keywords=Super+Flower+SF-650P14P)

clean installation of Windows 8.1 EFI mode

I also tried:

- 450 Watt bequiet PSU

- without riser. the box was powered by the original PSU (dont use akitio's PSUand a powered riser at once - it could melt your board and graphic card)

- another thunderbolt cable

- another macbook

- any order of switching on the box and booting

This egpu setup isn't "plug and play" at all as stated in other posts, but maybe I just got a faulty box.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.