5 ms·

Cromwellian made the point that the difference between a Quadro and a Geforce, at one time, was just a jumper change. It was exactly the same hardware -- and ch

by trimbo 13y ago

Cromwellian made the point that the difference between a Quadro and a Geforce, at one time, was just a jumper change. It was exactly the same hardware -- and changing that jumper activated the card's ability to give you better wireframe performance and whatever supposed advantages Quadro had. And for a long time, the workstation class cards were all that Autodesk would certify their software against. Several years ago, we users finally convinced them to start certifying against Geforce cards so that we could use consumer class cards for their software and take the "hit" in wireframe performance.

raverbashing 13y ago
"that the difference between a Quadro and a Geforce, at one time, was just a jumper change. It was exactly the same hardware" Are you sure it was the same hardware? Sure, changing the jumper makes it identify itself as a Quadro. But: Were the GPUs binned the same? http://en.wikipedia.org/wiki/Product_binning http://en.wikipedia.org/wiki/Product_binning Were the thermal solutions the same? Were the memories the same? And by 'the same',same speed characteristics? Did the cards have the same power circuitry (damping capacitors, shielding, load capacity, etc) But yes, I am in favour of using consumer class cards for 3D Apps, I would leave the 'pros' for the top end or specialist apps (like the Quadros that have multiple monitor outputs)
- cromwellian 13y ago
  Consumer cards have multiple monitor outputs too. Frankly, the rest of the stuff is scaremongering. Consumer class GPUs run gaming loads which stress the thermal envelope far more than say, Autodesk or SolidWorks. It's all FUD to sell someone a piece of hardware at 300% margin, and cognitive dissonance from those people who consider themselves "Pros" to have paid for a placebo. With the exception of NVidia cards with GPGPU tuned architecture (e.g. double precision FP at lower penalty) there is precious little advantage a workstation class GPU has. It's the same mentality of "no one ever get fired for buying IBM" as if buying the "Pro" version nets you any real advantage.
  - raverbashing 13y ago
    "Consumer cards have multiple monitor outputs too" I mean something like this: http://www.nvidia.com/object/nvs-compare-product-specs.html http://www.nvidia.com/object/nvs-compare-product-specs.html "Oh but you can add another video card", not always, this is needed in some cases (a niche case, true) For the other aspects, your guess is as good as mine. Sure, games stress the card more than Autocad, but maybe things like Maya put more (or different) stresses on it I don't disagree with you, I would use a regular video card first to see if it meets my needs. Now, for CUDA or other calculation apps, I would go for the Pro video cards, specifically because I would have someone to complain if it didn't work the way it should. And that's usually why things cost more sometimes.

by trimbo 13y ago

raverbashing 13y ago

"that the difference between a Quadro and a Geforce, at one time, was just a jumper change. It was exactly the same hardware" Are you sure it was the same hardware? Sure, changing the jumper makes it identify itself as a Quadro. But: Were the GPUs binned the same? http://en.wikipedia.org/wiki/Product_binning http://en.wikipedia.org/wiki/Product_binning Were the thermal solutions the same? Were the memories the same? And by 'the same',same speed characteristics? Did the cards have the same power circuitry (damping capacitors, shielding, load capacity, etc) But yes, I am in favour of using consumer class cards for 3D Apps, I would leave the 'pros' for the top end or specialist apps (like the Quadros that have multiple monitor outputs)

cromwellian 13y ago
Consumer cards have multiple monitor outputs too. Frankly, the rest of the stuff is scaremongering. Consumer class GPUs run gaming loads which stress the thermal envelope far more than say, Autodesk or SolidWorks. It's all FUD to sell someone a piece of hardware at 300% margin, and cognitive dissonance from those people who consider themselves "Pros" to have paid for a placebo. With the exception of NVidia cards with GPGPU tuned architecture (e.g. double precision FP at lower penalty) there is precious little advantage a workstation class GPU has. It's the same mentality of "no one ever get fired for buying IBM" as if buying the "Pro" version nets you any real advantage.
- raverbashing 13y ago
  "Consumer cards have multiple monitor outputs too" I mean something like this: http://www.nvidia.com/object/nvs-compare-product-specs.html http://www.nvidia.com/object/nvs-compare-product-specs.html "Oh but you can add another video card", not always, this is needed in some cases (a niche case, true) For the other aspects, your guess is as good as mine. Sure, games stress the card more than Autocad, but maybe things like Maya put more (or different) stresses on it I don't disagree with you, I would use a regular video card first to see if it meets my needs. Now, for CUDA or other calculation apps, I would go for the Pro video cards, specifically because I would have someone to complain if it didn't work the way it should. And that's usually why things cost more sometimes.