US$349 Mantiz Venus TB3 enclosure/dock
 
Notifications
Clear all

US$349 Mantiz Venus TB3 enclosure/dock  

 of  31
  RSS

bene552
(@bene552)
Active Member
Joined: 3 years ago
 

Hi!

I am really looking forward to you posting pics of a working X1 Carbon 5gen + Mantiz eGPU setup. Please don't keep us waiting Smile

Thanks in advance!

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Manta
(@manta)
Active Member
Joined: 3 years ago
 

"news! I tried this Mantiz Venus enclosure with a 2015 MacBook Air using the Apple Thunderbolt 3 (USB-C) to Thunderbolt adapter"

That's great news itsage. Can't wait to hear the details of the group buy.

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 
Posted by: bene552

 

Hi!

I am really looking forward to you posting pics of a working X1 Carbon 5gen + Mantiz eGPU setup. Please don't keep us waiting Smile

Thanks in advance!

   

There's no problem at all, I'll do it on Monday, the Venus is in our office.

 

Mantiz: ● ●


ikir liked
ReplyQuote
bene552
(@bene552)
Active Member
Joined: 3 years ago
 
Posted by: Mymantiz_John

 

Posted by: bene552

 

Hi!

I am really looking forward to you posting pics of a working X1 Carbon 5gen + Mantiz eGPU setup. Please don't keep us waiting Smile

Thanks in advance!

   

There's no problem at all, I'll do it on Monday, the Venus is in our office.

 

   

Great to hear that! If this is possible, could you also post some benchmarks for the X1C2017+ eGPU setup?
And also a different question: will the Mantiz Core be only available in gray? Or is a black version also planned

 

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
bene552
(@bene552)
Active Member
Joined: 3 years ago
 

I meant: Mantiz Venus, not Mantiz Core Smile

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Dsntrgr8
(@dsntrgr8)
Eminent Member
Joined: 3 years ago
 

Does it work well with the Razer Blade Stealth? I'd definitely get this instead of the Razer Core 

I'm thinking about getting a gtx 1050 for whatever egpu I get

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
nando4
(@nando4)
Noble Member Admin
Joined: 3 years ago
 

@all, some updates about this enclosure that occurred today:

Review, group buy links and comparison to other enclosures is at:

External GPU Buyer’s Guide: Top 2017 eGPU enclosures compared

eGPU.io Review: Mantiz Venus Review: All-in-one Thunderbolt 3 eGPU Docking Station

Group Buy Offer: [North America & Europe] Mantiz Venus Thunderbolt 3 eGPU Enclosure Group Buy

 

eGPU Setup 1.35    •    eGPU Port Bandwidth Reference Table

2015 15" Dell Precision 7510 (6th,4C,H) + GTX 1080 Ti @32Gbps-M2 (ADT-Link R43SG) + Win10 // compares M.2 vs TB3 performance


benr, itsage and ikir liked
ReplyQuote
benr
 benr
(@benr)
Eminent Member
Joined: 3 years ago
 

Does anyone know why none of the newer eGPU enclosures (including the Mantiz) are putting a second Thunderbolt 3 port on their designs for daisy chaining of other devices?

Is it just for the cost-savings (I appreciate it's a competitive market), or is there a technical reason to not do so?

Since there are no TB3/TB2 hubs/port multipliers in existence, it's a limitation I wish someone would overcome.

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
XCVG
 XCVG
(@xcvg)
Trusted Member
Joined: 3 years ago
 
Posted by: benr

 

Does anyone know why none of the newer eGPU enclosures (including the Mantiz) are putting a second Thunderbolt 3 port on their designs for daisy chaining of other devices?

Is it just for the cost-savings (I appreciate it's a competitive market), or is there a technical reason to not do so?

Since there are no TB3/TB2 hubs/port multipliers in existence, it's a limitation I wish someone would overcome.

   

I suspect they're running the USB and LAN ports off the passthrough, using it as USB3.1 connected to an internal hub. I can't confirm this, however.

Late-2015 15" Dell XPS 9550 GTX960M + [email protected] (Mantiz Venus) + Win10


ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 
Posted by: XCVG

 

Posted by: benr

 

Does anyone know why none of the newer eGPU enclosures (including the Mantiz) are putting a second Thunderbolt 3 port on their designs for daisy chaining of other devices?

Is it just for the cost-savings (I appreciate it's a competitive market), or is there a technical reason to not do so?

Since there are no TB3/TB2 hubs/port multipliers in existence, it's a limitation I wish someone would overcome.

   

I suspect they're running the USB and LAN ports off the passthrough, using it as USB3.1 connected to an internal hub. I can't confirm this, however.

   

HI. TBT3 chip is with two Lanes. we had use both of these two lanes, to have Daisy chain TBT3 output, we need to add the 2nd TBT3 chip on the board. Since we had already multi ports in the IO cards, we gave up this idea for the cost saving.

 

Mantiz: ● ●


ReplyQuote
titanmaru
(@titanmaru)
New Member
Joined: 3 years ago
 

I would love to see the EVGA 980ti hybrid water cooling system placement inside the Venus as the 120mm intake is coming from the bottom but the Node is coming from the front.  If anyone could provide picture would be highly appreciated because I have that card laying around, I believe @itsage has one as well.  If you could kindly test out for me that would be great!

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
BlackJack84
(@blackjack84)
Eminent Member
Joined: 3 years ago
 

@itsage, as the Mantiz Venus is equipped with a TI83 chip, have you already checked if its CUDA-Z H2D bandwidth is the same as the Node?

2016 12.5" Razer Blade Stealth (i7-7500U) + GTX1070 @ 32Gbps-TB3 (Mantiz Venus)


ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 

We are doing several experiments.

1. We have both version including the Ti65982 & TI 65983.

2. We have both version firmware including the eGPU & PCI-E, both are official version of the Intel's.

3. We will try to use the TI65983 board but load the " Previous FW " and unselect the " EGPU"  in the FW cooking option to let it be the PCI-e to see the difference.

I'll update the the status to you guys after we accomplish the test.

TI65983 VS TI65982 is nothing related with the issue, but Intel's FW for 82 & 83 are different, we are clarifying the FW which version to cause the difference then ask for Intel's opinion.

 

 

Mantiz: ● ●


itsage liked
ReplyQuote
itsage
(@itsage)
Illustrious Member Admin
Joined: 3 years ago
 
Posted by: titanmaru

I would love to see the EVGA 980ti hybrid water cooling system placement inside the Venus as the 120mm intake is coming from the bottom but the Node is coming from the front.  If anyone could provide picture would be highly appreciated because I have that card laying around, I believe @itsage has one as well.  If you could kindly test out for me that would be great!

I tried placing the GTX 980 Ti Hybrid inside this Mantiz Venus enclosure. It's a very tight fit. I don't recommend it though because the cooling lines would get kinked to get them fit.

Posted by: BlackJack84

@itsage, as the Mantiz Venus is equipped with a TI83 chip, have you already checked if its CUDA-Z H2D bandwidth is the same as the Node?

CUDA-Z numbers are similar to the AKiTiO Node. Per Mymantiz_John, TI83 controller's firmware is different than TI82. This may be the culprit. Hopefully, with people at AKiTiO and Mantiz looking into this, we'll have an answer soon.

Posted by: Mymantiz_John

We are doing several experiments.

1. We have both version including the Ti65982 & TI 65983.

2. We have both version firmware including the eGPU & PCI-E, both are official version of the Intel's.

3. We will try to use the TI65983 board but load the " Previous FW " and unselect the " EGPU"  in the FW cooking option to let it be the PCI-e to see the difference.

I'll update the the status to you guys after we accomplish the test.

TI65983 VS TI65982 is nothing related with the issue, but Intel's FW for 82 & 83 are different, we are clarifying the FW which version to cause the difference then ask for Intel's opinion.

We're looking forwards to seeing your updates.

external graphics card builds
best laptops for external GPU
eGPU enclosure buyer's guide


titanmaru liked
ReplyQuote
ed_co
(@ed_co)
Reputable Member
Joined: 3 years ago
 
Posted by: Mymantiz_John

 

We are doing several experiments.

1. We have both version including the Ti65982 & TI 65983.

2. We have both version firmware including the eGPU & PCI-E, both are official version of the Intel's.

3. We will try to use the TI65983 board but load the " Previous FW " and unselect the " EGPU"  in the FW cooking option to let it be the PCI-e to see the difference.

I'll update the the status to you guys after we accomplish the test.

TI65983 VS TI65982 is nothing related with the issue, but Intel's FW for 82 & 83 are different, we are clarifying the FW which version to cause the difference then ask for Intel's opinion.

Really nice to know!!

mid-2017 Macbook Pro RP560 + MSI GTX 1080Ti Gaming [email protected] (Mantiz Venus) + macOS10.13 & Win10


ReplyQuote
ed_co
(@ed_co)
Reputable Member
Joined: 3 years ago
 

As some people are asking for things, like colours, etc... I would like to ask 2 simple questions, that maybe could be interesting for others too.

There are 2 small thing that could make the Matiz Venus the BEST contender of any TB3 eGPU enclosures:

1) Be able to charge 100W like the Asus one.

2) Be a little less wider (the Razer Core is like 2 inches less, maybe just one inch less than now could be enough, not as thin as the Core).

How hard could be to perform this changes? Specially the first one, that should be PERFECT for any macbook pro 15 inches owner could dream.

Thanks.

mid-2017 Macbook Pro RP560 + MSI GTX 1080Ti Gaming [email protected] (Mantiz Venus) + macOS10.13 & Win10


ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 
Posted by: ed_co

 

As some people are asking for things, like colours, etc... I would like to ask 2 simple questions, that maybe could be interesting for others too.

There are 2 small thing that could make the Matiz Venus the BEST contender of any TB3 eGPU enclosures:

1) Be able to charge 100W like the Asus one.

2) Be a little less wider (the Razer Core is like 2 inches less, maybe just one inch less than now could be enough, not as thin as the Core).

How hard could be to perform this changes? Specially the first one, that should be PERFECT for any macbook pro 15 inches owner could dream.

Thanks.

   

HI:

Thanks for giving us the valuable suggestions. actually those 2 concerns were considered while we just started design in.

1. Let's talk about the Power delivery. First, Thunderbolt 3 Alpine Ridge did support 100W PD, but the TI6598X is able to afford to set the GPIO to change the PD Walt. the reason we keep it on the 60W is considering the PSU power consumption and thermal in the mainboard. with 60W current PD, the temperature of the chip had been 60 degree more, estimated the 100W, it will reach to 80 degree or more which may cause the component liability issue. 2nd, we need to consider the efficiency of the charging. we need to consider the latest notebook design as well, KabbyLake U, Skylake U and others latest CPU, the system total PWR consumption is limited to 37-85W. for the 85W scenario, it has to use the internal GFX such as MacBook Pro 2016 AMD 455. it means we don't consider the 85 for this moment so decided to take 60W to have better system liability & advantage of the Power supply.

2. Dimm. that's why we have MZ01 to afford the larger Monster Card. we try not to set the eGPU looks like the M-ATX tower besides our beautiful MacBook Pro.

Thank you.

Mantiz: ● ●


ikir, nando4 and itsage liked
ReplyQuote
itsage
(@itsage)
Illustrious Member Admin
Joined: 3 years ago
 

I tried fitting different GPUs inside this Mantiz Venus. The Sapphire R9 Fury Nitro is one of the largest air cooled graphics cards. It barely fits and looks great with this enclosure.

external graphics card builds
best laptops for external GPU
eGPU enclosure buyer's guide


Kooki, nando4, Yukikaze and 2 people liked
ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 
Posted by: ed_co

 

Posted by: Mymantiz_John

 

We are doing several experiments.

1. We have both version including the Ti65982 & TI 65983.

2. We have both version firmware including the eGPU & PCI-E, both are official version of the Intel's.

3. We will try to use the TI65983 board but load the " Previous FW " and unselect the " EGPU"  in the FW cooking option to let it be the PCI-e to see the difference.

I'll update the the status to you guys after we accomplish the test.

TI65983 VS TI65982 is nothing related with the issue, but Intel's FW for 82 & 83 are different, we are clarifying the FW which version to cause the difference then ask for Intel's opinion.

Really nice to know!!

   

HI all:

  Here's the result:

Please check the chart

this is related with the FW, we had asked Intel how this happened and why the FW caused such differences.

Mantiz: ● ●


ikir liked
ReplyQuote
Yukikaze
(@yukikaze)
Prominent Member Moderator
Joined: 3 years ago
 

That top result looks an awful lot like the reduced performance host to device that Node users are reporting. If I had to guess, the culprit is likely the same for both enclosures and has to do something with the FW provided by Intel.

Want to output [email protected] out of an old system on the cheap? Read here.
Give your Node Pro a second Thunderbolt3 controller for reliable peripherals by re-using a TB3 dock (~50$).

"Always listen to experts. They'll tell you what can't be done, and why. Then do it."- Robert A. Heinlein, "Time Enough for Love."


ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 
Posted by: Yukikaze

 

That top result looks an awful lot like the reduced performance host to device that Node users are reporting. If I had to guess, the culprit is likely the same for both enclosures and has to do something with the FW provided by Intel.

HI:

I would like to clarify an important concept here.

first. if we are talking about the EGPU , not a storage. actually the " Performance " won't be impacted by the H-D Value. Performance is the D-H ( after computed, calculated package ) send back to Host.

Bench mark utility may prove for this. But if we are talking about the PCI-E device which may use for the M.2 SSD, of course the  eGPU FW won't be right for this case.  

we had asked Intel what the purpose the FW is different in PCI-E & EGPU. but after measurement, we see nothing difference in external GFX " Performance ". Thank you.

Mantiz: ● ●


ReplyQuote
ddqp
 ddqp
(@ddqp)
Eminent Member
Joined: 3 years ago
 

waaal!

it's a good back to back test and a clear explanation of H-D and D-H. I should have bought your product!

Late Macbook Pro 2016 13' touch bar + AKITIO node + GTX 1050TI 4G Windows


ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 
Posted by: ddqp

 

waaal!

it's a good back to back test and a clear explanation of H-D and D-H. I should buy your product!

HI:

Then join the Group buy. 😀

Mantiz: ● ●


ReplyQuote
bene552
(@bene552)
Active Member
Joined: 3 years ago
 
Posted by: Mymantiz_John

 

Posted by: bene552

 

Hi!

I am really looking forward to you posting pics of a working X1 Carbon 5gen + Mantiz eGPU setup. Please don't keep us waiting Smile

Thanks in advance!

   

There's no problem at all, I'll do it on Monday, the Venus is in our office.

 

   

@Mymantiz_John

Just a polite reminder on posting the X1C5 + Mantiz pics / tests Smile Mantiz working well with the Carbon is an important factor for me when deciding on both the laptop and the enclosure. I would be really grateful if you could post some update on the issue - thank you in advance! Smile

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 
Posted by: bene552

@Mymantiz_John

Just a polite reminder on posting the X1C5 + Mantiz pics / tests Smile Mantiz working well with the Carbon is an important factor for me when deciding on both the laptop and the enclosure. I would be really grateful if you could post some update on the issue - thank you in advance! Smile

 

Please see below:
 

 

Mantiz: ● ●


itsage liked
ReplyQuote
bene552
(@bene552)
Active Member
Joined: 3 years ago
 
Posted by: Mymantiz_John

 HI, dunno what it went wrong with upload the pictures.

  

Thank you for running the tests and uploading the pics! Could you also post the 3dMark interpretation of the result, sth as here:  http://www.3dmark.com/3dm11/6739198

Thanks in advance!

 

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 

http://www.3dmark.com/3dm11/12084778

 

Mantiz: ● ●


ReplyQuote
nando4
(@nando4)
Noble Member Admin
Joined: 3 years ago
 
Posted by: Mymantiz_John

 

HI:

I would like to clarify an important concept here.

first. if we are talking about the EGPU , not a storage. actually the " Performance " won't be impacted by the H-D Value. Performance is the D-H ( after computed, calculated package ) send back to Host.

 

Mantiz Venus also delivering half H2D bandwidth like the AKiTiO Node

 

@John, Pls continue asking Intel to correct this problem. As is now, the Mantiz Venus with the half H2D TB3 bandwidth  problem, (1158MiB/s) will deliver less gaming FPS than a TB2 system (~1400MiB/s) when using an external LCD!!

It and the also TI83-based AKITiO Node having the same problem:  ALERT: AKITiO Node half-H2D CUDA-Z performance issue

 

The H2D bandwidth is critical in providing good FPS. It indicates how fast the TB3 channel is in sending data from the CPU to the eGPU. Traffic such as games texture maps. That is the primary traffic direction when using an external monitor attached to the eGPU.

The D2H becomes important when using Optimus or X-connect accelerated internal LCD mode where display traffic is send back from the eGPU to the iGPU.

3dmark benchmarks aren't bandwidth heavy enough to pick up this H2D anomoly.

Above: Late-2016 13" Macbook Pro + GTX980@TB3-32Gbps (Mantiz Venus) + macOS [itsage]

 

eGPU Setup 1.35    •    eGPU Port Bandwidth Reference Table

2015 15" Dell Precision 7510 (6th,4C,H) + GTX 1080 Ti @32Gbps-M2 (ADT-Link R43SG) + Win10 // compares M.2 vs TB3 performance


ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 
Posted by: nando4

 

The H2D bandwidth is in absolutely critical in providing good FPS. It indicates how fast the TB3 channel is in sending data from the CPU to the eGPU. Traffic such as games texture maps. That is the primary traffic direction when using an external monitor attached to the eGPU.

The D2H becomes important when using Optimus or X-connect accelerated internal LCD mode where display traffic is send back from the eGPU to the iGPU.

3dmark benchmarks aren't bandwidth heavy enough to pick up this H2D anomoly.

   

HI: I did some experiments to see the internal VS external display performance in Mantiz Venus (TI 83 H-D 12Gb ) VS Razer Core ( TI 82 H-D 21Gb )

aaa
BBB

I saw no difference in Unigine Heaven Benchmark in FPS.

 

Mantiz: ● ●


ikir and itsage liked
ReplyQuote
nando4
(@nando4)
Noble Member Admin
Joined: 3 years ago
 
Posted by: Mymantiz_John

 HI: I did some experiments to see the internal VS external display performance in Mantiz Venus (TI 83 H-D 12Gb ) VS Razer Core ( TI 82 H-D 21Gb )

I saw no difference in Unigine Heaven Benchmark in FPS.

   

As mentioned in my last post, 3dmark ( and now Heaven) are not heavy bandwidth apps. Internal vs external benchmarking is irrelevant. It's using the D2H bandwidth, which isn't affected.

Try some games that stress the H2D bandwidth. eg: Far Cry Primal that saw almost  a 15FPS drop (89->74) going from x4 3.0 (32Gbps) to x4 2.0 (16Gbps) here, which is equivalent to the half H2D problem we are seeing here.

Needless to say, the TI82-based Razer Core isn't affected but the TI83-based Mantiz Venus, AKiTiO Node and probably XG Station 2 are. Customers who are being sold a Mantiz TB3 enclosure are expecting TB3 throughput.

I can see Mantiz Venus purchase disputes or refunds down the track if this problem isn't fixed before enclosure delivery.

 

eGPU Setup 1.35    •    eGPU Port Bandwidth Reference Table

2015 15" Dell Precision 7510 (6th,4C,H) + GTX 1080 Ti @32Gbps-M2 (ADT-Link R43SG) + Win10 // compares M.2 vs TB3 performance


itsage liked
ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 

Thanks for your valuable suggestions and comment. I will test the game FPS both in Razer core and Venus tomorrow. 

Mantiz: ● ●


itsage, ikir and nando4 liked
ReplyQuote
bene552
(@bene552)
Active Member
Joined: 3 years ago
 
Posted by: Mymantiz_John

 

http://www.3dmark.com/3dm11/12084778

 

   

Thanks!

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
nando4
(@nando4)
Noble Member Admin
Joined: 3 years ago
 

@John, here is the also-affected AKiTiO Node attached to a  TB2 Macbook via the Apple TB3->TB2 adapter. I gets a H2D number that's higher than on native TB3 systems but is what's expected from TB2.

This does then say the half-H2D issue is some handshaking issue on native TB3 systems.

eGPU Setup 1.35    •    eGPU Port Bandwidth Reference Table

2015 15" Dell Precision 7510 (6th,4C,H) + GTX 1080 Ti @32Gbps-M2 (ADT-Link R43SG) + Win10 // compares M.2 vs TB3 performance


ReplyQuote
nando4
(@nando4)
Noble Member Admin
Joined: 3 years ago
 

@John, Matlab benchmarking results matching CUDA-Z results have been posted at the following URL. This being done to verify CUDA-Z results:

https://egpu.io/forums/thunderbolt-enclosures/alert-akitio-node-half-h2d-bandwidth-issue/#post-5541

eGPU Setup 1.35    •    eGPU Port Bandwidth Reference Table

2015 15" Dell Precision 7510 (6th,4C,H) + GTX 1080 Ti @32Gbps-M2 (ADT-Link R43SG) + Win10 // compares M.2 vs TB3 performance


ReplyQuote
Mymantiz_John
(@mymantiz_john)
Vendor
Joined: 3 years ago
 
Posted by: nando4

 

@John, Matlab benchmarking results matching CUDA-Z, to ensure CUDA-Z isn't buggy,  have been added to the following post:

https://egpu.io/forums/thunderbolt-enclosures/alert-akitio-node-half-h2d-bandwidth-issue/#post-5541

   

HI, Never doubt that CUDA Z goes wrong. let's see the result of Far Cry Primal FPS and also expect to hear the reply from Intel soon.

 

Mantiz: ● ●


ReplyQuote
Dsntrgr8
(@dsntrgr8)
Eminent Member
Joined: 3 years ago
 

If you were to put an hdd and a gpu in it, would the hdd bottleneck the gpu at all? For example, if I ran steam from the hdd inside the egpu, would that effect anything?

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
 of  31