nvidia NVIDIA Controls AIB Launch and Driver Distribution - HardOCP

[Taf the Ghost]

13 minutes ago, S w a t s o n said:

Considering the AIBs were still accepting massive shipments of 10 series cards and will be selling them alongside 20 series used 10 series cards are going to end up DIRT CHEAP by the time Ampere rolls around

I was thinking, originally, that Ampere was probably early 2020, but I'm pretty sure Nvidia is going to milk Turing for 2 full years, at this point. We'll get Volta v2, whatever it's called, in probably mid-2019 with maybe the first HBM3 on it. Nvidia will be more than happy to charge 20k USD a pop for those.

 

Actually, I honestly think we're going to get Icelake-S from Intel before Nvidia's 7nm consumer GPUs show up, and I'm not certain Icelake-S is ever launching.

[S w a t s o n]

17 minutes ago, Taf the Ghost said:

Actually, I honestly think we're going to get Icelake-S from Intel before Nvidia's 7nm consumer GPUs show up, and I'm not certain Icelake-S is ever launching.

I blame Raja, unless he redeems himself over at intel, i'll buy an intel gpu in a heartbeat if it's competitive

 

Edit: Should point I know that GCN's failures arent due to Raja but he still kind of fucked us over with Vega/prim shaders and just being misleading (Fury anyone?) in general

[Taf the Ghost]

5 minutes ago, S w a t s o n said:

I blame Raja, unless he redeems himself over at intel, i'll buy an intel gpu in a heartbeat if it's competitive

 

Edit: Should point I know that GCN's failures arent due to Raja but he still kind of fucked us over with Vega/prim shaders and just being misleading (Fury anyone?) in general

I think it's best to blame RTG's Marketing Department. They seemed really determined to not capitalize on where the products were always going to end up being. Less "Poor Volta" and far more "Freedom to Play" or stuff like that. And sorting out their distribution model for GPUs. AMD, even without the GPU Mining, ends up being insanely expensive in certain regions. It's best not to wipe out entire markets when they'd be more than able to buy up 560s & 570s. (If you had them to sell, but that's beyond the point.)

[S w a t s o n]

1 minute ago, Taf the Ghost said:

I think it's best to blame RTG's Marketing Department. They seemed really determined to not capitalize on where the products were always going to end up being. Less "Poor Volta" and far more "Freedom to Play" or stuff like that. And sorting out their distribution model for GPUs. AMD, even without the GPU Mining, ends up being insanely expensive in certain regions. It's best not to wipe out entire markets when they'd be more than able to buy up 560s & 570s. (If you had them to sell, but that's beyond the point.)

Well intel took Chris Hook so they can be misleading over at intel now, business as usual

[Taf the Ghost]

2 minutes ago, S w a t s o n said:

Well intel took Chris Hook so they can be misleading over at intel now, business as usual

Weird thing is, in the "gaming" space, Intel's marketing is actually quite good. The place their money in esports tournaments all over the place, make sure to get high visibility and all that jazz. Their entry into the dGPU market is probably going to go horribly.

 

And if their first GPU is a dud, you can already predict the "Blue Balls" memes.

[cj09beira]

7 hours ago, mr moose said:

If you are really worried about this then just wait till benchmarks from non approved reviewers come out.  Nvidia can't hide shit performance when they are selling their product to literally anyone who can walk into a shop and exchange cash for one.

 

Given they know this and every reviewer out there does too,  I don't think they are trying to curb reviews. Their NDA is no different from anyone else's so it's not likely some grand conspiracy. 

 

4 hours ago, Arika S said:

holy shit the tin foil hats in this thread.

 

If the cards suck, the benchmarks will come out. all they are doing is controlling who gets review samples. they cannot block "unapproved" reviewers from writing their own reviews.

 

oh wait, that doesn't go along with the current agenda..uhh, Nvidia evil blahahghqahgahgh

I believe they are aiming to control day 1 reviews, which unless you are GN you can't get your hands on a card unless they send you one, and those reviews make a huge difference to the public perception of the product

2 hours ago, ZacoAttaco said:

Gamers Nexus released a news video, basically NVIDIA still has so many 1000 series cards after the mining bubble burst that the plan is to have them sell alongside the new 2000 series RTX cards. He mentioned the different price range will appeal to different audiences and that the GTX name will still be used in the future.

 

I'm glad that for the time being at least the 2000 series isn't the only option. Unless the performance is insane, I wouldn't be interested in the 2000 series, simply too expensive so it's cool to see the 1000 series will still be around.

so the better course of action is to not buy either and wait for the prices to get lower, understood 

 

3 minutes ago, Taf the Ghost said:

I think it's best to blame RTG's Marketing Department. They seemed really determined to not capitalize on where the products were always going to end up being. Less "Poor Volta" and far more "Freedom to Play" or stuff like that. And sorting out their distribution model for GPUs. AMD, even without the GPU Mining, ends up being insanely expensive in certain regions. It's best not to wipe out entire markets when they'd be more than able to buy up 560s & 570s. (If you had them to sell, but that's beyond the point.)

wasn't poor volta a very old ad, when it was made there is a good chance they still hadn't discovered that ngg wasn't going to work properly and so they would have lower performance than expected, that one really back fired  

[mr moose]

45 minutes ago, cj09beira said:

 

I believe they are aiming to control day 1 reviews, which unless you are GN you can't get your hands on a card unless they send you one, and those reviews make a huge difference to the public perception of the product

 

To what end?  Let's play out that scenario:

 

nvidia control the reviewers for day one launch.  Why? does their product not perform as it should as people are claiming this is about? 

 

Answer 1. yes it performs like all other benchmarks that come after release,  Nothing is lost, they haven't changed anything, their "interference" has resulted in nothing.

 

Answer2. No, their product is an under performer and they want good reviews to avoid a drop in initial sales,  people buy the cards under false reviews (reviews nvidia cannot claim they had no hand in), people then find out they don't perform as they were presented and Nvidia gets their arse sued off like they did for the 970.   Do you think that is a good long term plan for them?   Having to face court costs and reimburse all effected customers?  I would bet a fair sum that if they are trying to control the reviewers because of the performance of the card then they will end up in court.

 

There is no logic to it being about performance. 

 

[cj09beira]

3 minutes ago, mr moose said:

To what end?  Let's play out that scenario:

 

nvidia control the reviewers for day one launch.  Why? does their product not perform as it should as people are claiming this is about? 

 

Answer 1. yes it performs like all other benchmarks that come after release,  Nothing is lost, they haven't changed anything, their "interference" has resulted in nothing.

 

Answer2. No, their product is an under performer and they want good reviews to avoid a drop in initial sales,  people buy the cards under false reviews (reviews nvidia cannot claim they had no hand in), people then find out they don't perform as they were presented and Nvidia gets their arse sued off like they did for the 970.   Do you think that is a good long term plan for them?   Having to face court costs and reimburse all effected customers?  I would bet a fair sum that if they are trying to control the reviewers because of the performance of the card then they will end up in court.

 

There is no logic to it being about performance. 

 

i just think they are trying to control the day 1 reviews, what for i have no clue, it could be that they want to make sure the benches will focus on RT, maybe they don't want the price compared against the 10 series, or there are some games they want to avoid/add 

[mr moose]

1 minute ago, cj09beira said:

i just think they are trying to control the day 1 reviews, what for i have no clue, it could be that they want to make sure the benches will focus on RT, maybe they don't want the price compared against the 10 series, or there are some games they want to avoid/add 

 

I don't think they have the power to do that. Too many reviewers to keep quiet.  An act like this can be considered antitrust, it would only take one going to the FTC and then all NDA's, Contracts, and emails between nvidia and AIB's/reviewers would be subpeana'd.  It wouldn't necessarily even take a reviewer to start this either, it could only take a handful of disgruntled consumers to complain.   I get why people would considered the GPP to be something, that was allegedly only between Nvidia and a handful of AIB partners, but this goes beyond AIB's and to reviewers/media as well.

[Lathlaer]

I'm shooting in the dark here but wouldn't this be a way for them to enforce their NDA that some signed and some didn't? 

 

For instance, let's assume news outlet X has not signed this NDA = they are not eligible for anything NVIDIA sends them. But how else is NVIDIA going to enforce it so that they don't get their GPU, say, from ASUS or from MSI or from Gigabyte? 

 

I think they are just trying to make sure that whoever blew their NDA doesn't get a Strix for review and thus completely circumventing the need to sign it. Realistically speaking, I don't think AIB partners have first hand knowledge about who signed the NDA and who didn't so they are trying to make sure that nobody who didn't sign doesn't get the card as review sample.

 

As @mr moose said, it would be completely unrealistic of them to try to "control" the content reviews because in this day and age that would only blow up in their faces. Yea, GPU are expensive but they are not THAT expensive so that a news outlet can't afford to go and buy $2500 worth of GPUs. Even if they do, the content they make with those GPU's will make them enough money to cover that expense and then some.

 

I may be wrong but this was my first instict.

[Athan Immortal]

17 minutes ago, Lathlaer said:

But how else is NVIDIA going to enforce it so that they don't get their GPU, say, from ASUS or from MSI or from Gigabyte? 

 

I think they are just trying to make sure that whoever blew their NDA doesn't get a Strix for review and thus completely circumventing the need to sign it. Realistically speaking, I don't think AIB partners have first hand knowledge about who signed the NDA and who didn't so they are trying to make sure that nobody who didn't sign doesn't get the card as review sample.

 

This is exactly it. As the HardOCP article mentions, in the past AIB partners were left to their own marketing, but now Nvidia is making sure that they can't just send those cards out willy nilly. They're not allowed to include the drivers, you have to get them from Nvidia, for which you need to sign the NDA or tough luck.

[spartaman64]

6 hours ago, Arika S said:

holy shit the tin foil hats in this thread.

 

If the cards suck, the benchmarks will come out. all they are doing is controlling who gets review samples. they cannot block "unapproved" reviewers from writing their own reviews.

 

oh wait, that doesn't go along with the current agenda..uhh, Nvidia evil blahahghqahgahgh

You think the average consumer would know which publishers signed the NDA and which didn't. They will do a Google search rtx 2080 reviews and click on the first one (hint it will probably be one of the ones published first) oh look Tom's hardware says the rtx 2080 is amazing and it's 500% faster well I got to just buy it 

[Arika]

21 minutes ago, spartaman64 said:

You think the average consumer would know which publishers signed the NDA and which didn't. They will do a Google search rtx 2080 reviews and click on the first one (hint it will probably be one of the ones published first) oh look Tom's hardware says the rtx 2080 is amazing and it's 500% faster well I got to just buy it 

there's a difference between choosing who can receive review samples and forcing reviewers to publish outright false information.

[leadeater]

4 minutes ago, Arika S said:

there's a difference between choosing who can receive review samples and forcing reviewers to publish outright false information.

Information doesn't have to be false to be misleading:

 

[spartaman64]

3 minutes ago, Arika S said:

there's a difference between choosing who can receive review samples and forcing reviewers to publish outright false information.

Why go through the effort of grabbing and lifting someone's legs to make them dance when you can just shoot at their feet

[Arika]

1 minute ago, leadeater said:

Information doesn't have to be false to be misleading:

 

 

Quote

oh look Tom's hardware says the rtx 2080 is amazing and it's 500% faster

outright false information, which is what i was responding to.

 

 

Just now, spartaman64 said:

Why go through the effort of grabbing and lifting someone's legs to make them dance when you can just shoot at their feet

I have no idea what point you're trying to convey with that statement.

[leadeater]

10 minutes ago, Arika S said:

outright false information, which is what i was responding to.

Well RTX 2080 Ti could be 500% faster, in specific RTX scenarios tailored to show off the differences and reviewed in just the right way to highlight that. But yes I get your point but there also isn't nothing to be gained by having tighter control over review samples, primarily those reviews will be first and the most viewed and drive the most initial sales.

 

Wouldn't it suck as a buyer if the reviews got you excited for the product and you went out brought a RTX 2080 because the reviews showed it doing much better than a 1080 Ti in the majority of tested games to then 1-3 weeks later say a Hardware Unboxed review came out with 35 games tested and it actually shows that the 1080 Ti is faster and cheaper, then you find out that few games will have any meaningful RTX features for some time and some may not have much at all when they do.

 

Buying the RTX 2080 wouldn't be that much of an issue but it also wouldn't make you feel any less mislead in that situation, you may have still gone for the RTX 2080 anyway but a fully informed purchase is better than a misinformed purchase even without a negative impact to the buyer.

[SpaceGhostC2C]

Calm down, people. It's just a move to ensure full transparency, so gamers know what they're buying.

[Arika]

10 minutes ago, leadeater said:

-snip-

having read the article again, it doesn't make the insinuation that the reviewer has to publish what Nvidia tells them to and only able to use the specifial games and benchmarks where it has the biggest increase, it's just about who has access to the review samples and drivers and normal NDA.

 

If there was something telling them "we will moderate the content you publish before it's made public to remove anything negative about our product" then HardOCP would absolutely have said that, and if that is the case, then no self respecting reviewer or review site would sign the NDA and anyone that does is not likely going to be a respected reviewer currently anyway.

 

 

As a buyer, I always wait multiple weeks to months after a product release to make sure there are enough reviews saturating the market with a good mix of positive and negative things. Hell people have already pre-ordered it even without any reviews already. those people are idiots.

[DimasRMDO]

That's so shady man....

[Brooksie359]

7 hours ago, leadeater said:

Not a $100 role. And current ram prices effects Pascal just as much as it effects Turing, new doesn't mean the price must go up, performance increase doesn't mean the price must go up, you should never expect GPU prices to go up just because it's new, something must justify it and it must be over and above the last generation and current factors. Turing isn't in a vacuum, it's in the same environment as everything else is. Either Turing was a significant development cost increase over Pascal and Volta, unlikely, or Nvidia put the price up just to align it with current street prices so it looks less bad.

 

Not that I care about the price increase, I gave you the why people are complaining. Doesn't have to make sense to you that's just why, price tag went up so naturally people are going to complain. People will always find something to complain about, be it prices being much higher than MSRP or the MSRP increasing, or performance increase lower than expected, literally anything can get people complaining.

The die size is much larger. The 2080ti is something like 750mm2 while the 1080ti is around 450mm2. The price to manufacture these chips went up because they increased cuda cores while adding both rtx cores and tensor cores. 

[TigerHawk]

I don't really get the point of controlling the drivers. As soon as they send them out to a reviewer, they'll be up on a torrent or FTP somewhere within 24 hours to share with other reviewers.

[79wjd]

12 minutes ago, TigerHawk said:

I don't really get the point of controlling the drivers. As soon as they send them out to a reviewer, they'll be up on a torrent or FTP somewhere within 24 hours to share with other reviewers.

Nvidia could very easily (and should) place a signature in the driver -- this would allow Nvidia to determine who uploaded the driver and then ban that reviewer from future releases.

[Tech Enthusiast]

11 hours ago, mr moose said:

If you are really worried about this then just wait till benchmarks from non approved reviewers come out.  Nvidia can't hide shit performance when they are selling their product to literally anyone who can walk into a shop and exchange cash for one.

 

Given they know this and every reviewer out there does too,  I don't think they are trying to curb reviews. Their NDA is no different from anyone else's so it's not likely some grand conspiracy. 

Exactly.

 

I am pretty certain this is due to the nature of the new tech.

Nvidia knows the new tech needs to be taken into consideration to reach the masses. Too many "i want moar FPS only!" - freaks out there.

I would imagine they just want to make sure the first reviews are done by people that really look at the whole picture and don't just test FPS and ignore all the added stuff.

 

Everyone can watch those boring/useless reviews a week later, so what is the matter with the conspiracy and tinfoil hattery again, seriously.

[pas008]

39 minutes ago, 79wjd said:

Nvidia could very easily (and should) place a signature in the driver -- this would allow Nvidia to determine who uploaded the driver and then ban that reviewer from future releases.

yes they want accountability of leaks too

20 minutes ago, Rattenmann said:

Exactly.

 

I am pretty certain this is due to the nature of the new tech.

Nvidia knows the new tech needs to be taken into consideration to reach the masses. Too many "i want moar FPS only!" - freaks out there.

I would imagine they just want to make sure the first reviews are done by people that really look at the whole picture and don't just test FPS and ignore all the added stuff.

 

Everyone can watch those boring/useless reviews a week later, so what is the matter with the conspiracy and tinfoil hattery again, seriously.

exactly they want to show the performance of their new tech along with its normal performance

and accountability

seems like no one think accountability isnt anything more especially using wrong drivers and leaks

 

leaks can hurt purchases good or bad leaks