NVIDIA forbids deployment of GeForce drivers in data centers...

[Angry Badger]

Not long ago, NVIDIA released an update to their end-user license agreement, and some of the additions to the agreement are rather alarming...

 

From TechPowerUp:

Quote

There's one particular statement in the limitations section that caught our eye. And it reads: No Datacenter Deployment. The SOFTWARE is not licensed for datacenter deployment, except that blockchain processing in a datacenter is permitted.

 

The recent addition can be found in the agreement which you have to accept before directly downloading GeForce-drivers from their site as the fourth point under section 2.1.3.

 

I find it disturbing that NVIDIA feels like they have to prohibit usage in data centers just because they fail to offer enough value to their customers through their insanely expensive Quadro and Tesla cards. Are NVIDIA admitting here that their enterprise solutions aren't good enough/offer enough value compared to their "mainstream" alternatives?

[Godlygamer23]

Pretty dumb to be honest. 

[kerradeph]

Quote

except that blockchain processing in a datacenter is permitted.

That part bugs me because they are basically saying "Yeah, feel free to keep running your mining farms which are utterly fucking over the prices".

[Godlygamer23]

Just now, kerradeph said:

That part bugs me because they are basically saying "Yeah, feel free to keep running your mining farms which are utterly fucking over the prices".

The whole thing is stupid.

[Angry Badger]

2 minutes ago, kerradeph said:

That part bugs me because they are basically saying "Yeah, feel free to keep running your mining farms which are utterly fucking over the prices".

Indeed, and thery're not even trying to hide their agenda a little bit. They're litterally telling every single enterprise customer out there that you're not allowed to buy the cheaper solutions.

 

Though it's NVIDIA telling ther customers what they can and can't do with their product which is the scary part in my opinion.

[Dabombinable]

6 minutes ago, Godlygamer23 said:

The whole thing is stupid.

Can they even enforce the EULA and dictate what drivers the owners can and can't use?

[Theguywhobea]

I think they just don't want people using consumer grade hardware in industrial scenarios like a data center. Their issue is if they don't explicitly forbid people from using consumer grade cards in their data centers, they will end up getting all kinds of support calls and issues where the customer believes this is actually Nvidia's fault where it's really their own fault for using the wrong hardware. 

 

To give a bit of an example, you wouldn't slap a netgear switch in a panel in a boiler room at school. Consumer grade routers aren't rated to operate in ambient temperatures that you might see in a metal box in a boiler room, but you can bet that the System Intergrator that installed that switch is going to have an unhappy customer when they come in in the morning to a cold building because their controllers aren't talking anymore.

 

Nvidia doesn't validate their consumer grade cards to run flat out 24/7, and their drivers aren't validated to have high accuracy when doing computations. The cost premium on Quadro cards come from the extra validation Nvidia does on the cards and the drivers.

[kerradeph]

1 minute ago, Raxzzer said:

Indeed, and thery're not even trying to hide their agenda a little bit. They're litterally telling every single enterprise customer out there that you're not allowed to buy the cheaper solutions.

 

Though it's NVIDIA telling ther customers what they can and can't do with their product which is the scary part in my opinion.

Part of it might also be that they are getting an unusual amount of RMA requests for devices being used in datacenters because the cards are installed in a matter which causes them to basically suffocate. The datacenter style graphics cards are designed with the high flow and pressure front to back airflow built into servers. This is just playing the benefit of the doubt that they're more interested in both selling the right product for the value of it and so that there aren't as many devices possibly being taken out of use due to improper deployment. 

[Matu20]

Another day, another stupid decision, monopoly is a bad thing.

[kerradeph]

4 minutes ago, Theguywhobea said:

Nvidia doesn't validate their consumer grade cards to run flat out 24/7, and their drivers aren't validated to have high accuracy when doing computations. The cost premium on Quadro cards come from the extra validation Nvidia does on the cards and the drivers.

I was thinking this too, but their allowance of using them in a datacenter deployment for mining means they're confident that they can do 24/7. If they're trying to be kind to the user, I would imagine it's for cooling issues. Airflow routes in a server is significantly different from the airflow in a computer case and this might be causing a lot of cards failing prematurely. 

[kerradeph]

1 minute ago, Matu20 said:

Another day, another stupid decision, monopoly is a bad thing.

Kind of hoping that Intel comes in and starts nipping at their heels at least in some parts of the market. AMD kind of fell on their face with their offerings.

[Enderman]

 

[Angry Badger]

12 minutes ago, Theguywhobea said:

I think they just don't want people using consumer grade hardware in industrial scenarios like a data center. Their issue is if they don't explicitly forbid people from using consumer grade cards in their data centers, they will end up getting all kinds of support calls and issues where the customer believes this is actually Nvidia's fault where it's really their own fault for using the wrong hardware. 

 

To give a bit of an example, you wouldn't slap a netgear switch in a panel in a boiler room at school. Consumer grade routers aren't rated to operate in ambient temperatures that you might see in a metal box in a boiler room, but you can bet that the System Intergrator that installed that switch is going to have an unhappy customer when they come in in the morning to a cold building because their controllers aren't talking anymore.

 

Nvidia doesn't validate their consumer grade cards to run flat out 24/7, and their drivers aren't validated to have high accuracy when doing computations. The cost premium on Quadro cards come from the extra validation Nvidia does on the cards and the drivers.

I don't think we can completely dismiss the fact that Tesla and Quadro cards are unreasonably priced compared to how they perform when compared to what you can get in the more mainstream segment. There are lots of workloads where a Titan X for example will do a very good job for a much more reasonable price, most notably when working with neural networks. This is a workload where Tesla cards are an extremely expensive option when you instead can fill the racks with cards that are much cheaper and perform just as needed, despite having a higher failure rate than their enterprise counterparts.

[Godlygamer23]

Can you get around this by using out-of-date drivers? Since the agreement doesn't apply to the older ones. The application is obviously limited, but I think for a time, you might be able to get around it.

[Princess Luna]

4 minutes ago, Raxzzer said:

Titan X

Wasn't this that prosumer card jack of all trades? I know lots of people who invest on one of those above any Quadro solution because it actually was good value there, I personally see this shutting down even further the TITAN X segment...

 

I have always felt that nVidia and pretty much all the others have always charged too much for the "certifications" for real if you do a product that should work doing something why the hell you get to overprice for the certainty it will work? it just feels like milking extra on the work they should do to all.

[AresKrieger]

Is this in a EULA because if it is I really don't care, they have about as much weight as a helium filled balloon, kinda funny Nvidia thought no one would bash them for it though.

 

Additionally this is just an aggro way of saying we do not support this if you do this and have an issue go f*@k yourself

[Theguywhobea]

10 minutes ago, Raxzzer said:

I don't think we can completely dismiss the fact that Tesla and Quadro cards are unreasonably priced compared to how they perform when compared to what you can get in the more mainstream segment. There are lots of workloads where a Titan X for example will do a very good job for a much more reasonable price, most notably when working with neural networks. This is a workload where Tesla cards are an extremely expensive option when you instead can fill the racks with cards that are much cheaper and perform just as needed, despite having a higher failure rate than their enterprise counterparts.

It's not like this line in the ToC says Nvidia will send in a SWAT team if they see you using a Geforce card in your datacenter, it's just because they must be tired of getting support calls from uneducated owners who thought they were saving a buck. This way they can deny these people the support they don't deserve. The price premium on Quadro cards is justified as I stated before. These cards are validated for data center use, the drivers are validated for data center use, and you get support for data center applications, none of those things are free, so they front load the cost onto the card. This is something that is repeated in almost every market. I work for a distributor, our prices in a lot of cases don't beat our competitors, but our customers are more than willing to pay more for their parts because they know we have knowledgle and experienced tech support staff who will provide them support for "free"

[NumLock21]

Their workstation class drivers are certified. Geforce ones aren't. I don't get all these stupid bashing on Nvidia.

 

 

[leadeater]

11 minutes ago, NumLock21 said:

Their workstation class drivers are certified. Geforce ones aren't. I don't get all these stupid bashing on Nvidia.

The issue is for a very large variety of workloads GeForce cards are just as fast as Quadros or Teslas so many people have opted for those cards saving huge amounts of money, likely also getting better performance by being able to buy a much faster GPU.

 

Nvidia didn't like this so the solution, rather than making more affordable products, is to disallow the use of GeForce cards in servers so yea plenty justification to have a bitch in my opinion.

[Matu20]

53 minutes ago, kerradeph said:

Kind of hoping that Intel comes in and starts nipping at their heels at least in some parts of the market. AMD kind of fell on their face with their offerings.

Yep, exactly.

[XenosTech]

1 minute ago, leadeater said:

The issue is for a very large variety of workloads GeForce cards are just as fast as Quadros or Teslas so many people have opted for those cards saving huge amounts of money, likely also getting better performance by being able to buy a much faster GPU.

 

Nvidia didn't like this so the solution, rather than making more affordable products, is to disallow the use of GeForce cards in servers so yea plenty justification to have a bitch in my opinion.

That reminds me of the time I walked into an architect firm and they all had 780's in their systems at the time instead of quadros/teslas.

[sazrocks]

What even qualifies as a datacenter anyways?

[NumLock21]

8 minutes ago, leadeater said:

The issue is for a very large variety of workloads GeForce cards are just as fast as Quadros or Teslas so many people have opted for those cards saving huge amounts of money, likely also getting better performance by being able to buy a much faster GPU.

 

Nvidia didn't like this so the solution, rather than making more affordable products, is to disallow the use of GeForce cards in servers so yea plenty justification to have a bitch in my opinion.

It's not like the drivers will refuse to install. Your on your own, when it comes to tech support.

[Zodiark1593]

6 minutes ago, XenosTech said:

That reminds me of the time I walked into an architect firm and they all had 780's in their systems at the time instead of quadros/teslas.

Well, a workstation isn't necessarily a datacenter. Heck, I use Blender a lot with my own GeForce card, and it seems as happy as can be, minus the 2GB RAM bit.

[Yoinkerman]

1 hour ago, Raxzzer said:

Not long ago, NVIDIA released an update to their end-user license agreement, and some of the additions to the agreement are rather alarming...

 

From TechPowerUp:

 

The recent addition can be found in the agreement which you have to accept before directly downloading GeForce-drivers from their site as the fourth point under section 2.1.3.

 

I find it disturbing that NVIDIA feels like they have to prohibit usage in data centers just because they fail to offer enough value to their customers through their insanely expensive Quadro and Tesla cards. Are NVIDIA admitting here that their enterprise solutions aren't good enough/offer enough value compared to their "mainstream" alternatives?

You're also not allowed to keep a local copy for installing or use gfe on more than one computer in your house