> The reason AMD has been tight lipped about the entire issue is because they cannot be seen enforcing retailer MSRPs (which would be illegal); they can however (((''suggest''))) the optimal price and then try to enable it using incentives like rebates
>Remember, it is illegal for AMD to force a price point on retailers. Our source believed that the market forces should naturally converge at a universal price point of $599
>Vegas cost structure is going to come down sometime in Q4 2017 when Hynix’s HBM2 memory enters the scene along with custom AIB variants
http://wccftech.com/radeon-rx-vega-64-msrp-pricing-update/
>>62053114
>amdfags will still defend this
>>62053114
And so the real truth of it all comes out
Gibbo, the lying cunt, just wants more shekels.
>580 almost 400 EUR
>56 at 399 EUR
>64 at 499 EUR
Suprise. With prices high from mining it should be expected that you don't get them at msrp.
>>62053114
Oh, also, here's how AMD fix this mess:
Sell it themselves, through a website of their own.
Seriously, I understand that limits them and they shouldn't ONLY sell it through themselves, but they could easily sort it out through Amazon for examples sake, so they sell it directly from AMD-RTG and set their own prices.
>>62053114
>WCCF tech
garbage. they were the ones who started the whole AMD fake launch prices bullshit
i spit on wccftech's grave
>>62053475
This is exactly what Nvidia did with FE. Founders Edition wasn't a cheap ploy for Nvidia to whore profits, it was a quick and dirty way to put price pressure on AIBs and faggot retailers who aimed to extort consumers. FE was available on Nvidia's website for some time. So, you're right anon.. But this would require business people and marketing groups to actually do their job.
> mfw when these morons let this shit fester for weeks until replying showing how worthless human beings are in such roles now-a-days.
I wouldn't pay $400 for vega64, who honestly cares about this shit?
>>62053490
wccftech was right about pic related since february.
>>62053490
>they were the ones who started the whole AMD fake launch prices bullshit
No, that was Gibbo, the owner of OcUK who also started the 70-100MH/s rumour.
>>62053534
That's exactly what I was thinking of outright referencing in my post, but wasn't 100% sure if it was correct or not due to not looking too far into it at the time.
If this is the only issue causing the price integrity concerns, and nothing to do with manufacturing costs suddenly jumping, then it would only make sense.
I've just tweeted @ Radeon & AMD, but you know that nothing will come of it. Still, worth a try.
>>62053534
But the FE was a 100€ more expensive. The whole FE bullshit was just to cash on the fact that for a month there wouldn't be aftermarket coolers or really any other way to get the card and AMD couldn't do shit about it.
>>62053553
>No, that was Gibbo
You mean 'that was known liar Gibbo', right?
>>62053543
>would not even pay 400$ for a card that is as good / better than a 1080
wow what a jew you are
>>62053342
And he was claiming to introduce a system to let actual people get their hands on cards too.
>>62053707
How the Vega is better than GTX 1080????
>MSRP Pricing Update
>Manufacturer Suggested Retail Price Pricing
>Price Pricing
>>62053114
>naturally converge at a universal price point of $599
$100 MORE THAN WHAT IT IS WORTH
FFS
Please AMD sell rx56 on steam one per customer
>>62054005
>most people don't even know what msrp means
woooowww
>>62053605
Go on camelcamelcamel or pcpartpicker or any price site. They stated the MSRP and then they build a custom package (FE) and stated the price of it. There was no bait and switch. Prices were stable for ~3 months.
Price were $50-$80 over MSRP not $100.
This is not the case w/ vega.
Furthemore, nvidia offered them on their website w/ zero markup beyond the price they published. It nothing like this shit show vega launch. The markup is north of $200, none are in stock and retailers/aibs are raping the shit out of consumers. Meanwhile Radeon is derping w/ this consistent (but we suggest MSRP and can do nothing about it). Nvidia did something.. They made a custom card and sold it on their fucking website at a price they promised for months which kept aib/retail pricing fixed.
This is factual and can be verified by any price history site.
Any commentary to the contrary is horseshit.
http://images.anandtech.com/doci/10336/GTX1070.jpg
$70 above MSRP
I paid the founders edition price for the top of the line OC'd AIB.
No AIB nor any retailer dared go any amount above the FE price for 3-6 months... because people could simply go to Nvidia's website and buy it for far cheaper.
AMD didn't do this. Their twiddling their thumbs playing pretend dummy while retailers/distributors/aibs rake their customers over coals.
Likely because they have shit for brains marketing/business staff who don't know how to stick it to the market and reign control over their distribution chain.
>>62055113
1080/1070 launch didn't have scalpers and miners grabbing every single part in stock with moments. There were some stock shortages that was sorted within a few months but not a huge demand so it wasn't that bad.
Also founders edition was never a custom card, it was the reference design with a built in early adopter tax with it's increased rrp.
>>62055113
You're retarded (which should go without saying for a frogposter really). There was no mining bubble when Pascal was released. Demand was a lot lower. The 1060 and 1070 are currently facing stock shortages and massive overpricing too. Any half-decent model of the latter is currently selling for $500. Based Nvidia, right? By your own logic it's even more fucking embarrassing not to have stock sorted and allow price inflation more than a year after launch than to sell out at launch.
Miners were always going to strip the stock and raise the price as soon as it was available, like they have every single other card that's good for mining. Stop being a fucking retard, retard.
>>62053534
>FE was available on Nvidia's website for some time.
They still are... I bought a couple of 1070s last month. Most of the time everything except the $1,200 Titan Xp is sold out, but if you keep your eye on it you can score the cards at non-inflated prices (limited to two cards per order).
>>62055113
do you avatarfag or something, cause i saw this image in an amd thread
>>62057837
Yes, he does. At least it's useful for knowing that the post was made by a complete retard.
>>62053114
If retailers are selling your product over MSRP, why wouldn't you sell them for more money in the first place?
You're essentially just letting them make free money, instead of making the free money yourself.
>>62053114
All i see is wait wait wait wait all fucking wait, what the fuck AMD when can I fucking properly buy your cards at standard price everything is fucking wait
i just want to build a new computer when will stock be replenished and cryptocurrency outlawed
>>62058575
>why wouldn't you sell them for more money in the first place?
because margins for AMD do not change, AMD gains NOTHING at all from this except moving less volume which hurts them
the most profitable for amd would be selling everything at MSRP but with the mining craze included for maximum volume
>>62053324
what are you poor?
>>62059563
You can't buy Nvidia cards at standard price either.
And why should retailers sell them at normal price to begin with? They can get away with $1-200 upsell on everything. Only a retard would sell the cards at MSRP now.
>>62059818
vega 64 and 1080 are 10% difference inpprice here
1080 is ~$800-850 for aib
64 is $850
but our prices are fucked from the beginning, actually seeing how vega performs without AA i'm somewhat tempted, difference is negligible in local currency except being reference
>>62059563
why are geese such shitheads
>>62059839
also on AA topic, did anyone test without AA?
I'd like to see how NCU performs without the throughput choke for science.
>>62059857
Performs exactly like how every card performs with no AA
It gains FPS but it still stays exactly in the same performance bracket
>>62059974
MSAA hurts vega more because it's choked, so the question is how fancy NCU performs without a choke compared to nvidia which has pretty good triangles throughput
The big joke is that HBM wasn't even needed. AMD's cards were not memory bottlenecked in the slightest.
Gibbo caught in a lie
I wana to punch that scumbag.
>>62060031
HBM is a prep work for MCM because it needs to be wide, very wide
pushing fabs to produce more hbm is a good thing, it will be cheaper in the long run
>>62059997
It stays in the exact same performance bracket
Gains some FPS but doesn't pull ahead of any cards
>>62054005
>retailer MSRPs
>retailer manufacturer suggested retail prices
>>62060103
vega is ahead to being with, why is everyone saying vega is slower?
it has problems with price sure, but it's not slower, the fuck?
it may not be 30% faster than 1080 but it's 3% faster, consistently, in dx12 and pretty much equal fps to fps in less favorable games like rotr
>>62060132
Because it had 14 months of extra development.
So rather than saying it's a piece of shit, it's easier to quantify as "slower".
>>62060204
again, how is this relevant if I want to buy GPU right now?
I'd get the point if nvidia was going tor release volta next month or even this year, but they won't
>>62060296
it's relevant *right now* because the 1080 does not currently have any supply issues. You can get one on-demand pretty much. A Vega 64, being new hardware and highly attractive to both gamers, miners and professional workloads, and ontop of that suffering supply issues, is going to be difficult to get at the advertised price point.
>>62060132
56 is slower than 1070 on average and the other is below the 1080 on average as well
That's not any bias, that's purely what every benchmark points to
This was posted the other day: https://www.pcgamesn.com/amd/amd-rx-vega-launch-price-reduction
>is our full intention of where we would suggest the product be priced. Not just for launch, but ongoing.
>ongoing
So MSRP was never raised or changed. Gibbo is just a lying cunt that fucked AMD over PRwise.
>>62060669
Not to mention he lied about this to make himself look good yet gouging prices higher at the same time.
>>62060666
>what every benchmark points to
says the words, but never posts the numbers
>>62060669
enjoy (google translate is your friend)
https://www.hardwareluxx.de/index.php/artikel/hardware/grafikkarten/44084-amd-radeon-rx-vega-56-und-vega-64-im-undervolting-test.html
>>62060778
fuck, wrong post sorry, it's meant for the one before...
>Quick note on primative shaders from my end: I had a chat with AMD PR a bit ago to clear up the earlier confusion. Primative shaders are definitely, absolutely, 100% not enabled in any current public drivers.
>The manual developer API is not ready, and the automatic feature to have the driver invoke them on its own is not enabled.
Oh wow.
Thanks Ryan.
>>62060532
sorry got tangled in local prices, I guess it makes sense to get 1080 everywhere else if it wasn't end of the cycle for nvidia
>>62060799
but why?...
>>62060810
Ha КУ 64 yжe зaвeзли?
>>62060818
Looks like development for gaymen drivers was started silly late.
Heck they finally fixed the power management in th recent update.
>>62060666
>666
>lying
Hello, satan.
>>62060819
nope, only FE
>>62060871
Fucking hell that sucks.
Sapphire Vega56 or at least 64 when.
I want to get one before dorra skyrockets, courtesy of пыня.
>>62059807
amd sells 1 vega 64 for 500$
amd sells 2 vega 64 for 1000$
amd sells 3 vega 64 for 3000$
I just solved all the miner issues, amd also can enforce msrp, and if partnered with someone like amazon could sell through amazon, possibly get a prime deal too.
>>62060880
same boat, radeon is best choice in case ruble drops again and nobody would be able to upgrade without selling kidneys, at least I'll have some peace of mind with it
>>62060799
Jesus.
>>62060890
>amd also can enforce msrp
that's against the law, anon, if they could we wouldn't have this mess to begin with
>>62060761
I clearly said on average buddy, improve your reading comprehension
Because you can cherry pick a few outliers means nothing
>>62060761
I average 90fps at 1440p with a 1070 hairworks off (which I assume is how this was benched)
>>62060911
by selling the cards themselves they can't enforce msrp?
>>62060921
Selling cards themselves is such a pain in the ass.
>>62060890
I don't think this is enforceable unless they sell them themselves they are not responsible for what vendors retailer's and end users do with them.
Long term will tell but if yields don't go up we can kiss Vega goodbye until a respin
>>62060926
The yields for Vega are more than okay, the node is mature.
It's just Vega10 is fighting for fab capacity with Zeppelin.
>>62060925
Based nvidia, doing the work nobody wants to do, keeping those damn scalpers in check
>>62060921
they don't have infrastructure for that, just think how much work it is to sell worldwide directly
they just offload cards to channel vendors which distribute them further, it's giant companies that do just that - distribute
>>62060941
Yes, by pricing them $100 higher for shitty shitblower with shitty PCB.
>>62060941
yeah, and still you couldn't buy pascal for a month because shortages, god it's been less than two weeks for vega, calm down
>>62060913
remove crysis3, and wow. it's suddenly much faster on average
>>62060957
I bought a 1070 about a week after aftermarket came around, and a 1060 FE first day for my other pc
>>62060992
Dirt 4 runs 60fps on medium settings on my fucking igpu laptop
Just throwing that out there
>>62060926
yea, amd cares what users do with their cards, see, when you take a 50+ million dollar loss because every single coin miner floods the market with your card and partners are telling you to fuck off with that shit they cant sell anything because the market is flooded, and you gain nothing in the market the cards were for (gaming) and the market that bought your cards just again, dicked you over so fucking hard you lose 50+ million dollars, they care.
get the cards in gamers hands, and they may be a recurring customer, especially with how amd drivers have been recently compared to nvidia, and all nvidia's bullshit they don't let you touch.
get the cards in the hands of miners again and you are looking at vega hitting 200-300$ the moment the market crashes due to stupidly large influx of cards.
sell the cards at an extreme markup to miners and you may recoup costs when the crash happens.
>>62060941
If I was amd, I would contract aibs to supply coolers for the gpus,
base card with blower - 500$
base card alone - 450-500$
cooler made by sapphire + 50-70$
possibly even cooler made by noctua or arctic
then cooler pre assembled +20-40$
There you go, good cooler on top of a warranty, and aibs wouldn't have room to bitch about you putting a good cooler on it (they do this to amd and intel where they threaten to fuck off making anything for their cpus if they make too good a stock cooler, amd has a good stock cooler only because of its necessity with bulldozer, but still couldn't put the wrath in a non partner computer because of their 'partners' threats)
with this amd would set msrp, not get dicked by retailers, get a bigger cut of the sales, and have a better warranty process (potentially) because they have the base cards there, and just stick different coolers on it in house.
Would be a nightmare to set up initially but would pay dividends in the long run.
>>62060953
Imagine if AMD was selling $499+100 Vega's from day one?
>>62060953
>having $100 more expensive as an option is bad
>having $100 more expensive forced on you is good
#teemred amirite
>>62060947
partner with a distributer and feed cost of using them to customers through shipping and handling. the shipping would be slightly higher than normal, but not back breakingly higher, at least not air cooled vega 64 costing 700-800$ currently for stock more.
>>62060992
remove amds stupidly good top 2 and 2 stupidly bad bottom (one of which is well known to be fucking shit on both cards) and you get a better average picture.
>>62060992
I'm glad I've had a $380 dual-fan 1070 to enjoy since July of last year
>>62061057
>well known to be shit on both cards
But one of them is practically 1/5 shittier
Will there ever be a iChill x4 Vega? I'd buy that for 1000+ dollars.
>>62061085
>But one of them is practically 1/5 shittier
No, its a literal crapshoot if the game will magically work on one computer or not, 2 computers with the exact same specs get wildly different results in that game along with random crashes because its barely functional, shit is beyond fucking broken. But because its a desperation genre (massive world battle royal) and literally no other game is close to done or even better then this pile, everyone who desperately wants it play it, and right now thats 500,000 people right now and 700,000 people in the last 24 hours.
broken from the get go games should not be on a benchmark period.
Dirt at least has an excuse of the game is finished, plays well on both cards but amd handles global illumination better,
>>62060992
I saw this article, the 1070 beats it like 3/4 tests, they just pick a single resolution where it pulls ahead and hypes the fuck out of that, while casting "yeah the 1070 pulls ahead by 4 frames but that's margin of error they're equal!"
Go look at the review, it's actually embarrassing how they spin it
>>62061094
asus looks not bad this time around, 2.5 slot coolers are my dream, i'd take even 3 slot with anti bend stick if it means it will be quieter
>>62061274
>1080p gets wrecked by 1070
>4K gets wrecked by 1070
>"our final thoughts about this game bench is that vega pulls ahead by nearly 5% in 1440p, a solid win for vega"
The review is fucking cringe but at least there's data, and it's nice to see that there is a trend of vega pulling ahead of a 1070 at higher resolutions
But I mean
That's literally what the Fury X was hyped over, this just has a little more substance to the claim
>>62061274
>single resolution
I'm sorry but I have only one,single 1440p monitor, so I care only about that resolution
>>62061324
You literally sound like them
>4K is unrealistic for the majority of PC gamers, and nobody buying vega is aiming for 1080p, so let's just focus on 1440p, where vega pulls ahead on average, and not on averages across all resolutions
>>62061315
Why this pcb so long? Isn't vega's pcb small?
>>62061317
And 5% ahead is 1/20th of the value
1/20th of $400 is $20
And a 1070 can be found for $380
Therefore they're not competing with each other
Buy both
>>62061350
what? why the fuck would I care about 1080p/4k if I don't use the resolution? for "lul 1070 is better" argument?
I see you are not in the market to buy a GPU. I got 390 exactly because it was better in 1440p than 970 by same margins btw, and probably will get vega 56 because it's better in 1440p(no there was no need for new gpu last year now there is).
>>62061393
You get Nano if you want smOll card.
>>62061397
I'm not trying to sell you a GPU I'm saying their review spin is fucking transparent and retarded and that you're defending it
I also only own 1440p monitors, it's still a shitty spin of a review
>>62061393
nah, vega pcb is normal size, it's pretty good PCB if nothing else
there is nano vega coming alter, who knows when
>>62061411
I have to be honest, I didn't even listen to what reviewer said just skimmed through graphs.
>>62061397
>>62061411
I've also had no need for a GPU since last year
Because I bought a 1070 for $380 in july
Waitfags get cucked on prices by miners or poor yields
https://forum.beyond3d.com/threads/amd-vega-hardware-reviews.60246/page-59#post-1997699
Silly RTG what are you doing.
>>62061422
Pretty sure i remember seeing reference vega with small pcb and radiator much longer than pcb.
>>62061439
That was a devboard for AiBs.
>>62061430
That's smart I should've just done that
This whole board could take a lesson from you
>>62061393
>save pcb space with HBM stacking
>has to use the gained space for VRM
>>62060761
Now try it undervolted on the newest drivers.
>>62061436
Boy, rustilng some jimmies over there with the culled polygon throughout graph proved productive.
>>62062173
Well done, Rasterizer-kun.
>>62062132
>1080ti: 74-91fps
Entire graph is nazi propaganda
Do not trust the fucking krauts
>>62062302
Sure thing.
>>62062302
Your graph shows a dip in the 70s and averages in the high 90s, that's not that much different to what the chart is saying.
>>62062680
>>62062302
Infact your graph is an overclocked aib, it's showing pretty much the same results for the 1080 as well. The nvidia results line up, it's the undervolt on Vegas that still needs more evidence.
>>62062416
funny that because a 1080 with a 7600 can score 5696 according to google. Thats stock too
>>62062416
>no name arabian site
>>62062756
6700 sorry
>>62062897
Sorry, that greasy haired fuck Steve hasn't posted his results on the new driver yet, so you'll have to live with the German results until then.
>>62062756
>funny that because a 1080 with a 7600 can score 5696 according to google. Thats stock too
Why did I waste my time?
>>62060992
This was run on pre undervolt drivers. It would be much higher than that now. Also Dirt 4 was runon CMAA instead of MSAA which is why it is so silly high. I can't wait for full feature drivers to be rested though.
>>62061411
The number of 4K 'gaming' monitor owners is non-existent and people are not going to buy a 56 for 1080 (well most won't). 1440P is the sweet spot RIGHT NOW so it's relevant.
>>62063938
what is vsr
what is dsr
>>62060132
Because you compare cords based on price. And based on price the 64 is not faster than a titan, so you either pay $50 more and get a titan, or pay $150 less and get a 1080.
>>62063938
1440 is a meme that needs to die. High refresh is worthless out side of single player due to low server ticks. 4k 60hz is the current sweet spot of gaming if you are using 1080ti or vega 64.
>>62066021
Wrong
>>62067027
Explain why you need 144hz when the server tick is only 24 times a second?
>>62067707
Not all servers have the same tickrate and not every game is multiplayer.
>>62067914
You seem to lack reading comprehension
>>62068178
Not all servers have the same tickrate, retard. Show some proof that all servers are at 24hz or fuck off.
>>62068199
the guy clearly said it was useless outside of single player, you're just a flaming faggot
>>62068238
It's not, you and the other retard simply don't understand how netcode works but that isn't surprising considering this is /g/
>>62068193
is this good or bad? 675 for a 1080 card?
>>62068193
>2 whole cards
wow its fucking nothing
>>62066021
>tfw gunning down airplanes in 16v16 matches in glorious 144hz ulmb on max settings 1440p
It's a thing of beauty
>>62068270
Very bad
>>62064033
You tell me.
>>62054795
but ppl will just register a new account and buy one by one
>sold out everywhere
Who cares.
>>62071395
Not if it's required to be a year old
>>62067707
Guess you don't need more than 24fps either then..
I swear, if the Vega pricing hysteria is this insane now, just imagine how much louder the wailing and gnashing of gaymur nerdrage will get once Vega 56 launches.
>>62072077
The blood will be on your hands
>>62072077
There's idiots posting business invoices as if it's proof of some grand conspiracy on AMD's part. For fucks sake how much worse could it get?
>>62068193
>>62072077
When does V56 launches?
>>62072147
The 28th I think, could be wrong though.
>>62072188
reference? hope they don't need 2 weeks for partnercards
>>62072686
Yes, reference Vega 56.
>The reason AMD has been tight lipped about the entire issue is because they cannot be seen enforcing retailer MSRPs
Why does AMD care about MSRP ?
The money AMD gets is not MSRP, it is whatever the wholesale price is to retailers.
The difference between whole sale and whatever the retailer sold it for , goes back into the retailers pocket not AMDs
>>62072768
fuck
vega ages like little children, more you use them the better they get
>>62053114
What is there even to do anymore about this