It's not just gamers that can't get Graphics cards, it's scientists as well.

[leadeater]

10 minutes ago, asus killer said:

they shouldn't be buying 500$ graphics cards in the 1st place, those are gamers cards, for the consumer gaming market. Different market from the scientific cards. I guess no one prohibits them from doing so, still, not the same market.

Not all researchers are flush with money and have set amounts with goals they need to achieve with it. Also if your task does not require a Quadro then why should you buy one? Just because you see the word research doesn't mean they need ECC VRAM or some obscure driver optimization.

 

You know who calls Quadros professional cards, Nvidia. Yep I'm sure they are totally impartial on the matter and have nothing to gain by up-selling you to a card you don't need.

 

Don't blame the researchers for putting RGB LEDs on GPUs, it might look 'unprofessional' but it's even more so to waste money on things you don't need.

[grimreeper132]

17 minutes ago, leadeater said:

Not all researchers are flush with money and have set amounts with goals they need to achieve with it. Also if your task does not require a Quadro then why should you buy one? Just because you see the word research doesn't mean they need ECC VRAM or some obscure driver optimization.

 

You know who calls Quadros professional cards, Nvidia. Yep I'm sure they are totally impartial on the matter and have nothing to gain by up-selling you to a card you don't need.

 

Don't blame the researchers for putting RGB LEDs on GPUs, it might look 'unprofessional' but it's even more so to waste money on things you don't need.

more I read this more I think of my PC which sorta is a work PC for my course, which has RGB on it as that was the cheapest RAM and GPU (with good cooling, I wanted a tri fan design rather than a dual for my 1080ti just so I had max cooling)

[TOMPPIX]

mining is exactly like selling cs go skins. huehuehhue

[asus killer]

46 minutes ago, SpaceGhostC2C said:

"shouldn't"? What does it even mean? They will buy whatever cover they needs at the lowest cost, like anyone else.

Those are GPUs, specialized processors focus on matrix operations. They are for the whoever can benefit from matrix handling market.

Except maybe you, I guess  

everyone has every right to buy whatever for whatever reason, it's a free world (for the most part), still i was just correcting you when you said it was "one market", it isn't and it's split by the manufacturers. I should have written "shouldn't" and not shouldn't, that part is my bad.

The point still stands they are buying something that was not intended for the use they are giving it. It may seem irrelevant but it isn't, for nvidia they have to calculate sales based in predictions, and if people are buying cards for uses and by users they never accounted for then shortages are more likely. And it's not by a few gpus, it's a deviation by the hundreds for just one case, some all up and you can see how it goes so wrong. A gamers buys one/two cards at most. A project like this buying cards are removing the cards intended/predicted by nvidia for hundreds of gamers.

 

That's my point.

[SpaceGhostC2C]

31 minutes ago, asus killer said:

everyone has every right to buy whatever for whatever reason, it's a free world (for the most part), still i was just correcting you when you said it was "one market",

 

Yes, it is, the market for GPUs is just one market in which GTX and Quadros (and non-Nvidia cards) are sold, but even if you want to treat market segmentation as two different markets (if it was, you wouldn't need segmentation strategies to begin with), then it would still be the case that the "GTX market" is just one market, and in that market buyers can be gamers, miners, scientists, deep-fakers, shiny object collector, or confused vaccum cleaner buyers. Whoever they are, they are all demand side in the same, single market for GTX GPUs, and when the price goes up, it goes up for everyone. 

The news would be if any group had some special deal or other form of shielding against price changes, compared to the general public.

 

[Coaxialgamer]

9 hours ago, BuckGup said:

AMD and Nvidia can easily produce more GPUs but they are scared the prices might drop and they would loose some money. Also Pascal is now end of life

Keep in mind that there still is a global DRAM shortage... 

[asus killer]

37 minutes ago, SpaceGhostC2C said:

Yes, it is, the market for GPUs is just one market in which GTX and Quadros (and non-Nvidia cards) are sold, but even if you want to treat market segmentation as two different markets (if it was, you wouldn't need segmentation strategies to begin with), then it would still be the case that the "GTX market" is just one market, and in that market buyers can be gamers, miners, scientists, deep-fakers, shiny object collector, or confused vaccum cleaner buyers. Whoever they are, they are all demand side in the same, single market for GTX GPUs, and when the price goes up, it goes up for everyone. 

The news would be if any group had some special deal or other form of shielding against price changes, compared to the general public.

 

 

"About NVIDIA
NVIDIA (Nasdaq: NVDA) is the world leader in visual computing technologies and the inventor of the GPU, a high-performance processor which generates breathtaking, interactive graphics on workstations, personal computers, game consoles and mobile devices. NVIDIA serves the entertainment and consumer market with its GeForce® graphics products, the professional design and visualization market with its Quadro® graphics products and the high-performance computing market with its Tesla™ computing solutions products. NVIDIA is headquartered in Santa Clara, Calif. and has offices throughout Asia, Europe and the Americas. For more information, visit www.nvidia.com."

http://www.nvidia.com/object/io_1238405290161.html

 

they even have separate divisions. Agree to disagree and lets move on. 

[dfsdfgfkjsefoiqzemnd]

5 hours ago, Bananasplit_00 said:

1: why not get quadros to begin with? 

I regularly watch the Puget build streams (not right now, they're on a hiatus when it comes to streaming), and plenty of the machines they build for scientific purposes have multiple 1080Tis in them instead of Titans or Quadros.  The difference in performance doesn't seem to justify the extra cost.

[Nicnac]

13 hours ago, LtStaffel said:

Is this a joke? Who needs GPUs to listen for extra terrestrial broadcasts, and better yet, what is their scientific evidence that such broadcasts have the potential to happen? What are the chances of it happening?

A grant sounds good to help scientists do really good stuff like find a cure for canc----no, they're listening to space with radios... Maybe the grant should go to folding@home instead?

+1

It sounds like the people runnig the SETI@home project. I suggest you take a read on their site as it's really interesting:

https://setiathome.berkeley.edu

 

Also, think of it more like looking for any sign of communication or signs of advanced technology, not something like a radio broadcast ^^

[grimreeper132]

1 hour ago, asus killer said:

 

"About NVIDIA
NVIDIA (Nasdaq: NVDA) is the world leader in visual computing technologies and the inventor of the GPU, a high-performance processor which generates breathtaking, interactive graphics on workstations, personal computers, game consoles and mobile devices. NVIDIA serves the entertainment and consumer market with its GeForce® graphics products, the professional design and visualization market with its Quadro® graphics products and the high-performance computing market with its Tesla™ computing solutions products. NVIDIA is headquartered in Santa Clara, Calif. and has offices throughout Asia, Europe and the Americas. For more information, visit www.nvidia.com."

http://www.nvidia.com/object/io_1238405290161.html

 

they even have separate divisions. Agree to disagree and lets move on. 

they might have diffrent divisions but the people who they sell to are who ever wants to buy them, they aren't picky as they can't be their a company and they sell to who ever.

 

As to why a scientist uses them, budgets, if they need the preofrmance of a 1080ti and they have the choice between a 1080ti eant to be £700 closer to £800-900 or a P6000 at £5,000 they both have near identical preformance, they will go 1080ti as it's alot cheaper, and it means they can get more, which considering that they seam to be buying about 64 at at time they seem to need, it saves them alot for the nearly the same profomance. Using these cards as a just a rendering a game is not using the full power of these cards, they are designed that they can do stuff like scientific research even if they aren't targeted at this. From the sounds of it they don't need the extra features from the more expensive cards either, as if they did they would probably buy them instead.

[ADM-Ntek]

not directly related but here a little something why space exploration and NASA are a good thing.

 

[grimreeper132]

Just now, ADM-Ntek said:

not directly related but here a little something why space exploration and NASA are a good thing.

 

I don't get how people think that NASA, ESA etc. are wastes of money they aren't they are pushing the boundaries of science and engineering which is never a bad thing and help us a hell of alot. Also it's really fucking cool, and I can't wait till we have a settlement on mars cause I really hope that happens in my lifetime.

[TVwazhere]

10 hours ago, BuckGup said:

But they could still ramp it up a bit. The demand has faded but they could have easily flooded the market a month ago and sitll made money. Still though people would buy GPUs

Quote

When asked about increasing card production, we were given a few primary answers:

• Betting on cryptocurrency is a big bet, because the mining market has not proven stable, and has proven unpredictable. To order on current demand doesn’t mean those cards show up tomorrow – they show up in a few months, and if cryptomining dies down in that time, that’s a big bag to be left holding. The manufacturers are forecasting months ahead (quarterly, actually), not weeks ahead.
• NVIDIA and AMD could likely produce more GPUs, but board partners need to actually place an order for them. We’ve seen some uninformed content creators online accuse nVidia and/or AMD of undersupplying the market. Well, nVidia and AMD need a customer to sell those GPUs to – that’d be the board partners. If they’re not ordering more, nVidia and AMD aren’t going to make more. Simple as that.
• New supply is showing up weekly, but it’s selling out fast. The best bet is to show up at a local retailer and ask when the next shipment comes in, then go there that day.
• There is concern about over-production, especially if mining falls enough that the second-hand market becomes flooded, outstripping the ability of first-parties to make money.

https://www.gamersnexus.net/industry/3211-what-do-manufacturers-think-of-mining-and-gpu-prices

[Zodiark1593]

3 hours ago, grimreeper132 said:

true but budgets don't always allow that, and if they need 64 cards about as powerful as a 1080ti (means to be about £650-700, are about £800-900) the cards they are meant to be buying are double to tripple that price, for not a massive gain in profromance they will take the consumer cards

If I was looking into building a large render server for Blender, Quadros would also be the very last cards I would look at too. The 1080 TI is faster in raw CUDA performance than the fastest Quadro (besides the GP100), and if that is what these scientists require, it would make no practical sense for them to consider going Quadro either.

[grimreeper132]

14 minutes ago, Zodiark1593 said:

If I was looking into building a large render server for Blender, Quadros would also be the very last cards I would look at too. The 1080 TI is faster in raw CUDA performance than the fastest Quadro (besides the GP100), and if that is what these scientists require, it would make no practical sense for them to consider going Quadro either.

That's pretty much what I am trying to say

[asus killer]

2 hours ago, grimreeper132 said:

they might have diffrent divisions but the people who they sell to are who ever wants to buy them, they aren't picky as they can't be their a company and they sell to who ever.

 

As to why a scientist uses them, budgets, if they need the preofrmance of a 1080ti and they have the choice between a 1080ti eant to be £700 closer to £800-900 or a P6000 at £5,000 they both have near identical preformance, they will go 1080ti as it's alot cheaper, and it means they can get more, which considering that they seam to be buying about 64 at at time they seem to need, it saves them alot for the nearly the same profomance. Using these cards as a just a rendering a game is not using the full power of these cards, they are designed that they can do stuff like scientific research even if they aren't targeted at this. From the sounds of it they don't need the extra features from the more expensive cards either, as if they did they would probably buy them instead.

money is not the issue, neither the types of cards. They don't want a graphics card to search for aliens, they want the same as miners, a calculator. What they want ideally is cheap rx570's same as miners. They could do it with cpus, with cheaper or with more expensive gpus, but it's a performance vs electricity cost.

I'm not an expert but if they really wanted a graphics card per se the quadros have different specifications, it's not like comparing video game performances.

The problem that i was point out is if everyone starts using consumer video cards for whatever calculations, be it mining, scientific or whatever then it's really hard for companies to predict sales and plan production accordingly, it helps me see the side of Nvidia and AMD. That's why they separate divisions, different buyers, different sales persons, etc... so they know the different markets.

[LtStaffel]

6 hours ago, miagisan said:

You are kidding right? Most astronomical "listening" and "observing" are almost never done using audio or visual observations. 

I was pointing out that extra terrestrial life probably does not affect us as much as other more immediate issues that GPUs could be used to enhance the research of.

[grimreeper132]

Just now, LtStaffel said:

I was pointing out that extra terrestrial life probably does not affect us as much as other more immediate issues that GPUs could be used to enhance the research of.

yea true, but thats the example given by the BBC for some reason

[Zodiark1593]

1 hour ago, asus killer said:

money is not the issue, neither the types of cards. They don't want a graphics card to search for aliens, they want the same as miners, a calculator. What they want ideally is cheap rx570's same as miners. They could do it with cpus, with cheaper or with more expensive gpus, but it's a performance vs electricity cost.

I'm not an expert but if they really wanted a graphics card per se the quadros have different specifications, it's not like comparing video game performances.

The problem that i was point out is if everyone starts using consumer video cards for whatever calculations, be it mining, scientific or whatever then it's really hard for companies to predict sales and plan production accordingly, it helps me see the side of Nvidia and AMD. That's why they separate divisions, different buyers, different sales persons, etc... so they know the different markets.

One could say that it was Nvidia that encourages such use in the first place by bringing CUDA to consumer cards. Many "gaming" cards even advertise CUDA capability, so in a sense, they are also marketed for "whatever calculations" as well as gaming.

[mynameisjuan]

15 hours ago, BuckGup said:

AMD and Nvidia can easily produce more GPUs but they are scared the prices might drop and they would loose some money. Also Pascal is now end of life

Well if you never noticed, but production kept up just fine before last year. Then cyrpto skyrocketed and demand shot up. So Nvidia and AMD have two options. Either spend millions ramping up productions on an already dead generation or spend those millions instead on their Volta production that is soon to be released. 

 

Its bullshit that people are blaming Nvidia and AMD over the shortage of cards. Its like people think companies can magically pull another factory out of their ass in no time with no cost to their bottom line. Especially for an end of life product. 

 

Not to mention the fact that even if they ramped up production, the demand would be higher than ever due to fact of TOR. I mean imagine if a 1080 still only cost $500 when people at the moment are willing to pay $1200+ for one. It would be an endless fucking cycle. 

[bcredeur97]

10 hours ago, mr moose said:

This reminds me of a few years back when a few scientists were explressing there concerns about the reserves of helium.  They were upset it was being wasted on party balloons when it was desperately need for research.  Being lighter than everything except hydrogen when released it floats of into the atmosphere from there it can actually escape gravity.

 

But fear not, just like Helium I am sure they will find a new natural reserve somewhere of 1080ti's or vega's. 

if they literally wait a month there will probably be thousands of 1080ti's available for cheap if Nvidia actually releases new cards (they probably will) that are significantly better than last gen.

[mr moose]

9 hours ago, leadeater said:

And they said I was crazy when I suggested they'd be mining for GPU's in the future.

[pyrojoe34]

23 hours ago, Canada EH said:

scientists can afford the premium and buy in bulk if they require lots.

Yea, that’s not true at all. Budgets are being slashed every year (thanks Congress, the cost of one of those tanks the army doesn’t even want could fund our lab for a decade) and most grants are given preferentially to big, well funded labs leaving the rest of us clawing for basic supplies and reagents.

 

Our molecular docking core runs simulations using 1080s and 1080Tis and we are perpetually short on computational power (the job queue is always growing). The core will buy a couple new ones whenever the budget allows and once a tower runs out of space (we run 8 per system), we kill several months of budget to get a new tower to add to the cluster.

 

Science is already hard enough when you’re well funded but regularly putting projects on standby because you run out of reagents or have to repair 20yo equipment yourself because replacing it is 6mo of budget and a tech coming out is half that, is a massive resource drain.

 

We have been the world leader in scientific research for 100+yrs but now we think that science is a waste of money unless there is a profit to be made and we’re falling further and further behind. If people want to make America great then maybe we should support the research that lays the foundation for every single tech that industry is able to use for innovation, progression, and prosperity.

 

It seems like half the people here think that researchers have a ton of money to spend? That's only true for maybe 5% of labs at the top few universities but the vast majority of academic labs are not limited by a lack of interesting research prospects or lack of breakthroughs, the number 1 limiting factor is budget. Every PI (PhD/Professor that runs a lab) I know spends the vast majority of their time writing grant proposals and seeking new funding sources and rely of grad students and lab techs to do the actual research. Ever wonder why professors seem to spend so little time concerned with the courses they teach and TAs do a lot of teaching? It's because they have spend all their time trying to fund the lab using a constantly shrinking pool of funding.

[kokakolia]

Errr...This is just a controversial thought: 

 

Just think about all of the cheap RX480/RX580/GTX 1060 flooding eBay once the hash rates of these cards drops significantly and miners can't unload these cards soon enough. It will be a glorious day for PC gaming! Well...RAM prices are still going up...so...

[kokakolia]

3 hours ago, pyrojoe34 said:

Yea, that’s not true at all. Budgets are being slashed every year (thanks Congress, the cost of one of those tanks the army doesn’t even want could fund our lab for a decade) and most grants are given preferentially to big, well funded labs leaving the rest of us clawing for basic supplies and reagents.

 

Our molecular docking core runs simulations using 1080s and 1080Tis and we are perpetually short on computational power (the job queue is always growing). The core will buy a couple new ones whenever the budget allows and once a tower runs out of space (we run 8 per system), we kill several months of budget to get a new tower to add to the cluster.

 

Science is already hard enough when you’re well funded but regularly putting projects on standby because you run out of reagents or have to repair 20yo equipment yourself because replacing it is 6mo of budget and a tech coming out is half that, is a massive resource drain.

 

We have been the world leader in scientific research for 100+yrs but are now we think that science is a waste of money unless there is a profit to be made and we’re falling further and further behind. If people want to make America great then maybe we should support the research that lays the foundation for every single tech that industry is able to use for innovation, progression, and prosperity.

 

It seems like half the people here think that researchers have a ton of money to spend? That's only true for maybe 5% of labs at the top few universities but the vast majority of academic labs are not limited by a lack of interesting research prospects or lack of breakthroughs, the number 1 limiting factor is budget. Every PI (PhD/Professor that runs a lab) I know spends the vast majority of their time writing grant proposals and seeking new funding sources and rely of grad students and lab techs to do the actual research. Ever wonder why professors seem to spend so little time concerned with the courses they teach and TAs do a lot of teaching? It's because they have spend all their time trying to fund the lab using a constantly shrinking pool of funding.

I understand what you're saying but as a former University student this infuriated me a lot. Tuition fees are blowing up and the quality of the Education couldn't be more disappointing! University professors didn't care much about teaching, but they were very passionate and proud about their research. I relied mostly on my TAs for everything. The professors we're marginally useful. You could always tell they were too busy with research to provide a positive learning environment. 

 

My point is Universities suck at teaching because they focus way more on research and funding over learning. In my experience, Tech School was wayyy better at teaching me things as the instructors were more involved with the students.