It is too expensive even at educational pricing level. Cheap nvidia pushing for profit even on such a low volume market as dev platforms... way to go.
Buying that for a desktop system? Even at 300$, you can get an ITX board, CPU and RAM and still get a marginally better system.
I was just about to compliment nvidia for making a dev kit that is not huge for once, then I noticed the "motherboard" it plugs into. So the "smaller than a credit card" bullet-point is just shallow marketing, as the SOC module doesn't seem to be usable on its own.
Also, I don't see OpenCL mentioned anywhere, so thanks but no thanks!
"So the "smaller than a credit card" bullet-point is just shallow marketing" nvidia may be the king for shallow marketing but this one isn't the huge motherboard is good for development phase but the final product should work without it
the probably bad news here is it looks like you must buy both of them as a package
It would have been better to put some of the headers on the soc board, and make the big board optional.
At any rate, I expect in a few months there will be something like a new ODROID board comparable in performance at like half the price of this one.
As impressive as the 1 TFlop number may seem, this is for FP16, so I don't expect more than 300 GFlops for FP32 and something entirelly pathetic for FP64. So in reality it is only slightly faster than competing GPUs from the same "generation", which should find their way into an ODROID or another similar ARM single board solution.
The base board is not really huge, but it is still too big to put on a typical robot or drone. Would have been nice to be able to use it standalone, but absent the headers and connectors this is not an option.
"It would have been better to put some of the headers on the soc board" I don't think they kept it without headers but I can't tell until I see the bottom of the SoC board because it looks like there where you connect the peripherals
"I don't expect more than 300 GFlops for FP32" actually it's 512 GFlops and forget about FP64, it's maxwell
"I expect in a few months there will be something like a new ODROID board comparable in performance at like half the price of this one." Agree, they are not alone here but I hop they make similar dev kit SoC board with "optional" big motherboard
Uh, I'm pretty sure the educational price isn't about using it as a simple desktop computer.
It's a dev kit, for use in robotics and other embedded development. It seems like something my university's embedded systems lab might buy for research.
The large motherboard for it is again just for development. Most small embedded systems have similar boards that you use for development and programming. The end product that you develop doesn't need all of that extra hardware so you can get by with something a lot smaller.
Yeah, the $299 educational price is good, but the $600 retail price will practically kill any momentum this things desperately needs in the corporate sector where commercial applications are developed. And that's there the money is. For something with a BOM of <$100, a 600% markup is downright unreasonable.
So in your expert opinion, 600$ are gonna be a problem for corporations who develop high tech computation intensive products?
Funny thing is, even though the price will not be problematic to any corporation, they could actually score more design wins if they gave those away on the cheap, possibly subsidize the price of those few dev platforms they are going to sell from the sales of the actual hardware for the retail products.
So what's the motivation? Why push for such a ridiculous profit margin on a dev platform, which is not a volume product, when it could hurt the actual volume sells later on?
That was kind of my point. They are being cocky charging so much for a devkit when similar kits have sold for <$100 from Intel and Google, and even Microsoft.
This isn't a PlayStation developer station or something proprietary where you license the "compiler" it's a platform preview kit if anything. There is no reason to milk it.
$600 is for the devkit (board plus breakout board). If you're purchasing breakout boards from Nvidia in bulk and putting the whole breakout board into your product, you're Doing It Wrong. The intended solution is the same as the RPi Compute Module: you develop with the development kit, and once you have your design working you fab a custom breakout board with only the parts you need (generally integrated into the PCB of your product) then drop in the bare TX1 module.
How is difference between this board and rooted Nvidia Shield, is there some big advantage? Because HW is same except 400 pin connector and price is 2 times bigger..
These are development boards, their purpose is for you to develop a software around the hardware on the board for some specific purpose, usually in embedded development. I believe some of their previous boards were used in cars, for image recognition and system control.
If you're happy with the peripheral hardware built into a shield and don't need access to any hardware debugging features a rooted shield is going to be better. ie If you're going to install a linux distro someone else ported and either use off the shelf software or only write high level software a dev board isn't for you. If you need to write custom low level kernel code in general; or need to interface with some external hardware that's not related to watching cat tube or playing angry birds readily accessible IO headers, hardware debug points, and official support for flashing a new firmware when something goes wrong and your device fails to boot is worth many times the price difference between a dev kit and rooted consumer hardware.
For people making medium sized hardware production runs splitting the dev kit into two parts, one with the core hardware, and a second with all the IO will offer significant hardware design savings. The most difficult/hard to make work right parts of building a system are on the small board with the CPU/ram/etc; designing a small IO board with the handful of ports you need is much less difficult work, and can be significantly cheaper to manufacture. Removing all the extra IO your application doesn't need also gives a free way to secure it against more casual tampering. (It won't keep out anyone who's determined to get in; but just denying USB/network/etc ports for them to plug cables into will stop a lot of people from trying.)
or just adding TX1 connector to an existing large board is
And just to add to that, NVIDIA tells us that the fan shouldn't even kick on under load in a normal room temperature environment. The giant HSF is more for dealing with heat in tight spaces if someone uses the dev kit in a stand-alone device.
R&D dude. Think of all those people who worked on this product. Had to be at least 25 people. Engineers, Technical Writers, Logistics, Managers, etc...
Each of those people have a salary, lets say $50,000 a year ( and thats low for this field). Additionally, money is needed to pay for their benifits such as health insurance, Dental, 401K)
Besides the employee's, Huge amounts of money is needed to tool the manufacturer for the designs, and find a distributer and pay for shipping to end user.
Add all this stuff up and the product launch cost millions. I didnt even mention the office space and utility bills these people use to make it.
Any product launch is like this.. When you quote some simple Bill of Materials vs Cost, you don't even come close figuring the profit margin.
So for basically $300-600 I can use this to make autonomous machines?
If for $1,000 I can upgrade my car to be self-driving, I think that's a pretty sweet deal. Can someone buy this and start a kickstarter campaign for automobile mod kits plz?!
It would be nice to know if this board follows some sort of an existing standard (like qseven, or smarc, etc.), to see an ecosystem of boards that this compute module plugs into, environmentally resistant cases, the list goes on; but be that as it may...
I COULD see these things catching on like literally wildfire, with people using them as the next hacker-whatever-thing like an arduino; But definitely NOT at these price points!
Its nice that this product class removes the abstraction layer of "Needing to know a guy", or "Establishing a relationship between your company & nvidia" for purposes of the uniting hackerspace & making bleeding edge widgits that no one ever thought of, but with the consumer end of this technology already being established, fully packaged, and priced at $199/ea (ShieldTV), the price point is REALLY going to kill this thing... I'd love to tear the stereo out of my truck, and drop one of these with an in dash monitor in its place and have my whole media collection (and depending on storage capacity, maps for the whole country, some videos, the list goes on -Maybe get REALLY crazy, and have it speak canbus to the whole vehicle & have new visuals for the gauges, etc.), THEN really get started with putting this on some quad/hex copters, and the list goes on, then move on to a litany of other embedded projects; But...
Yea, price point... It would be nice to see the dev/experimenter's/whatever kit at a max of $299 (the current educational price, and if so inclined -even more aggressive pricing for the educational market) & then the compute modules at ~$100'ish (in quantities of ONE... if you listen to the marketing guys, $99.99)
Would make it alot more feasible for joe-schmoe to have a pile of these at home for experimentation purposes
e-con Systems announces the launch of the much awaited MIPI camera board for NVIDIA® Jetson Tegra X1 development kit - e-CAM130_CUTX1. e-CAM130_CUTX1 is a 13MP 4 lane MIPI CSI-2 camera board based on AR1820 CMOS Image sensor from ON Semiconductor® and an integrated Advanced Image Signal Processor(ISP) for NVIDIA® Jetson Tegra X1 development kit. https://www.e-consystems.com/13mp-nvidia-jetson-tx...
e-con Systems announces the launch of the much awaited MIPI camera board for NVIDIA® Jetson Tegra X1 development kit - e-CAM30_CUTX1 - 3.4 MP Jetson TX1 Camera Board. The e-CAM30_CUTX1 is a 3.4 MP low light sensitive 4 lane MIPI CSI-2 camera board based on AR0330 CMOS image sensor from OnSemiconductor. The e-CAM130_CUTX1 streams 1080p(FullHD) @30fps, 3.4 MP@ 30 fps in uncompressed YUV format.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
34 Comments
Back to Article
p1esk - Tuesday, November 10, 2015 - link
I'm totally buying one of those as a linux desktop. No brainer at $299 educational pricing.ddriver - Wednesday, November 11, 2015 - link
It is too expensive even at educational pricing level. Cheap nvidia pushing for profit even on such a low volume market as dev platforms... way to go.Buying that for a desktop system? Even at 300$, you can get an ITX board, CPU and RAM and still get a marginally better system.
I was just about to compliment nvidia for making a dev kit that is not huge for once, then I noticed the "motherboard" it plugs into. So the "smaller than a credit card" bullet-point is just shallow marketing, as the SOC module doesn't seem to be usable on its own.
Also, I don't see OpenCL mentioned anywhere, so thanks but no thanks!
KTF26 - Wednesday, November 11, 2015 - link
"So the "smaller than a credit card" bullet-point is just shallow marketing"nvidia may be the king for shallow marketing
but this one isn't
the huge motherboard is good for development phase
but the final product should work without it
the probably bad news here is it looks like you must buy both of them as a package
ddriver - Wednesday, November 11, 2015 - link
It would have been better to put some of the headers on the soc board, and make the big board optional.At any rate, I expect in a few months there will be something like a new ODROID board comparable in performance at like half the price of this one.
As impressive as the 1 TFlop number may seem, this is for FP16, so I don't expect more than 300 GFlops for FP32 and something entirelly pathetic for FP64. So in reality it is only slightly faster than competing GPUs from the same "generation", which should find their way into an ODROID or another similar ARM single board solution.
The base board is not really huge, but it is still too big to put on a typical robot or drone. Would have been nice to be able to use it standalone, but absent the headers and connectors this is not an option.
KTF26 - Wednesday, November 11, 2015 - link
"It would have been better to put some of the headers on the soc board"I don't think they kept it without headers
but I can't tell until I see the bottom of the SoC board
because it looks like there where you connect the peripherals
"I don't expect more than 300 GFlops for FP32"
actually it's 512 GFlops
and forget about FP64, it's maxwell
"I expect in a few months there will be something like a new ODROID board comparable in performance at like half the price of this one."
Agree, they are not alone here
but I hop they make similar dev kit
SoC board with "optional" big motherboard
stetrain - Wednesday, November 11, 2015 - link
Uh, I'm pretty sure the educational price isn't about using it as a simple desktop computer.It's a dev kit, for use in robotics and other embedded development. It seems like something my university's embedded systems lab might buy for research.
The large motherboard for it is again just for development. Most small embedded systems have similar boards that you use for development and programming. The end product that you develop doesn't need all of that extra hardware so you can get by with something a lot smaller.
Samus - Wednesday, November 11, 2015 - link
Yeah, the $299 educational price is good, but the $600 retail price will practically kill any momentum this things desperately needs in the corporate sector where commercial applications are developed. And that's there the money is. For something with a BOM of <$100, a 600% markup is downright unreasonable.ddriver - Wednesday, November 11, 2015 - link
So in your expert opinion, 600$ are gonna be a problem for corporations who develop high tech computation intensive products?Funny thing is, even though the price will not be problematic to any corporation, they could actually score more design wins if they gave those away on the cheap, possibly subsidize the price of those few dev platforms they are going to sell from the sales of the actual hardware for the retail products.
So what's the motivation? Why push for such a ridiculous profit margin on a dev platform, which is not a volume product, when it could hurt the actual volume sells later on?
Samus - Thursday, November 12, 2015 - link
That was kind of my point. They are being cocky charging so much for a devkit when similar kits have sold for <$100 from Intel and Google, and even Microsoft.This isn't a PlayStation developer station or something proprietary where you license the "compiler" it's a platform preview kit if anything. There is no reason to milk it.
edzieba - Thursday, November 12, 2015 - link
$600 is for the devkit (board plus breakout board). If you're purchasing breakout boards from Nvidia in bulk and putting the whole breakout board into your product, you're Doing It Wrong.The intended solution is the same as the RPi Compute Module: you develop with the development kit, and once you have your design working you fab a custom breakout board with only the parts you need (generally integrated into the PCB of your product) then drop in the bare TX1 module.
ruthan - Tuesday, November 10, 2015 - link
How is difference between this board and rooted Nvidia Shield, is there some big advantage? Because HW is same except 400 pin connector and price is 2 times bigger..Braincruser - Tuesday, November 10, 2015 - link
These are development boards, their purpose is for you to develop a software around the hardware on the board for some specific purpose, usually in embedded development. I believe some of their previous boards were used in cars, for image recognition and system control.DanNeely - Tuesday, November 10, 2015 - link
If you're happy with the peripheral hardware built into a shield and don't need access to any hardware debugging features a rooted shield is going to be better. ie If you're going to install a linux distro someone else ported and either use off the shelf software or only write high level software a dev board isn't for you. If you need to write custom low level kernel code in general; or need to interface with some external hardware that's not related to watching cat tube or playing angry birds readily accessible IO headers, hardware debug points, and official support for flashing a new firmware when something goes wrong and your device fails to boot is worth many times the price difference between a dev kit and rooted consumer hardware.For people making medium sized hardware production runs splitting the dev kit into two parts, one with the core hardware, and a second with all the IO will offer significant hardware design savings. The most difficult/hard to make work right parts of building a system are on the small board with the CPU/ram/etc; designing a small IO board with the handful of ports you need is much less difficult work, and can be significantly cheaper to manufacture. Removing all the extra IO your application doesn't need also gives a free way to secure it against more casual tampering. (It won't keep out anyone who's determined to get in; but just denying USB/network/etc ports for them to plug cables into will stop a lot of people from trying.)
or just adding TX1 connector to an existing large board is
testbug00 - Tuesday, November 10, 2015 - link
Nvidia quoting 1TFLOP and not mentioning it's FP16. Terrific. Thanks Nvidia.There are standards which with specs are discussed, everyone uses FP32 for the most part for quoting GPUs.
Yojimbo - Tuesday, November 10, 2015 - link
It depends on the context. Here they are marketing it towards a segment with great interest in FP16 performance. It's not meant to play Fallout 4.testbug00 - Wednesday, November 11, 2015 - link
They didn't mention the FP16 on the slide. They obviously informed someone, or sites deduced it.Without it there you need to assume it's FP32.
HighTech4US - Wednesday, November 11, 2015 - link
assume: making an a** out of youMrSpadge - Wednesday, November 11, 2015 - link
Others would report FP16 performance if they were any faster at that than with FP32.ant6n - Tuesday, November 10, 2015 - link
All links in this article refer to other Anandtech articles.Yojimbo - Tuesday, November 10, 2015 - link
All whopping three of them.Yojimbo - Tuesday, November 10, 2015 - link
Maybe they should link to I Effing Love Science for diversity's sake.ToTTenTranz - Wednesday, November 11, 2015 - link
No OpenCL?HighTech4US - Wednesday, November 11, 2015 - link
Only AMDroids care about OpenCL. No one else does.Anato - Wednesday, November 11, 2015 - link
Last picture tells it a lot. Big heatsink and Fan. Ok for dev board but not so practical in small custom battery powered device.HighTech4US - Wednesday, November 11, 2015 - link
Quote: heatsink-fan for cooling (which we’re told is grossly overpowered for TX1)I'm sure charlie will take the picture and calculate that the X1 SOC uses 50-60 watts like he did with the original Jetson power brick.
Ryan Smith - Wednesday, November 11, 2015 - link
And just to add to that, NVIDIA tells us that the fan shouldn't even kick on under load in a normal room temperature environment. The giant HSF is more for dealing with heat in tight spaces if someone uses the dev kit in a stand-alone device.Shadowmaster625 - Wednesday, November 11, 2015 - link
How does $25 worth of silicon balloon into $600?HighTech4US - Wednesday, November 11, 2015 - link
Ask AppleMorawka - Saturday, November 14, 2015 - link
R&D dude. Think of all those people who worked on this product. Had to be at least 25 people. Engineers, Technical Writers, Logistics, Managers, etc...Each of those people have a salary, lets say $50,000 a year ( and thats low for this field). Additionally, money is needed to pay for their benifits such as health insurance, Dental, 401K)
Besides the employee's, Huge amounts of money is needed to tool the manufacturer for the designs, and find a distributer and pay for shipping to end user.
Add all this stuff up and the product launch cost millions. I didnt even mention the office space and utility bills these people use to make it.
Any product launch is like this.. When you quote some simple Bill of Materials vs Cost, you don't even come close figuring the profit margin.
kgardas - Wednesday, November 11, 2015 - link
~$300 is not bad for A57 together with 4GB RAM. IMHO the cheapest board in A57/A72 performance range. Commercial price is a little bit off probably...webdoctors - Wednesday, November 11, 2015 - link
So for basically $300-600 I can use this to make autonomous machines?If for $1,000 I can upgrade my car to be self-driving, I think that's a pretty sweet deal. Can someone buy this and start a kickstarter campaign for automobile mod kits plz?!
BiGG-D - Wednesday, November 11, 2015 - link
Hopefully nVuhDIA is reading these?Anyways...
Shield TV $199 (Quantity 1)
Jetson TX1 BSP $599/299 respectively (Quantity 1)
Jetson TX1 Modules $299 (Quantity 1000)
It would be nice to know if this board follows some sort of an existing standard (like qseven, or smarc, etc.), to see an ecosystem of boards that this compute module plugs into, environmentally resistant cases, the list goes on; but be that as it may...
I COULD see these things catching on like literally wildfire, with people using them as the next hacker-whatever-thing like an arduino; But definitely NOT at these price points!
Its nice that this product class removes the abstraction layer of "Needing to know a guy", or "Establishing a relationship between your company & nvidia" for purposes of the uniting hackerspace & making bleeding edge widgits that no one ever thought of, but with the consumer end of this technology already being established, fully packaged, and priced at $199/ea (ShieldTV), the price point is REALLY going to kill this thing... I'd love to tear the stereo out of my truck, and drop one of these with an in dash monitor in its place and have my whole media collection (and depending on storage capacity, maps for the whole country, some videos, the list goes on -Maybe get REALLY crazy, and have it speak canbus to the whole vehicle & have new visuals for the gauges, etc.), THEN really get started with putting this on some quad/hex copters, and the list goes on, then move on to a litany of other embedded projects; But...
Yea, price point... It would be nice to see the dev/experimenter's/whatever kit at a max of $299 (the current educational price, and if so inclined -even more aggressive pricing for the educational market) & then the compute modules at ~$100'ish (in quantities of ONE... if you listen to the marketing guys, $99.99)
Would make it alot more feasible for joe-schmoe to have a pile of these at home for experimentation purposes
But, I suppose we'll see... :/
econsystemss - Wednesday, January 18, 2017 - link
e-con Systems announces the launch of the much awaited MIPI camera board for NVIDIA® Jetson Tegra X1 development kit - e-CAM130_CUTX1. e-CAM130_CUTX1 is a 13MP 4 lane MIPI CSI-2 camera board based on AR1820 CMOS Image sensor from ON Semiconductor® and an integrated Advanced Image Signal Processor(ISP) for NVIDIA® Jetson Tegra X1 development kit.https://www.e-consystems.com/13mp-nvidia-jetson-tx...
econsystemss - Wednesday, March 22, 2017 - link
e-con Systems announces the launch of the much awaited MIPI camera board for NVIDIA® Jetson Tegra X1 development kit - e-CAM30_CUTX1 - 3.4 MP Jetson TX1 Camera Board. The e-CAM30_CUTX1 is a 3.4 MP low light sensitive 4 lane MIPI CSI-2 camera board based on AR0330 CMOS image sensor from OnSemiconductor. The e-CAM130_CUTX1 streams 1080p(FullHD) @30fps, 3.4 MP@ 30 fps in uncompressed YUV format.https://www.e-consystems.com/3MP-Jetson-TX1-Camera...