AMD's Initial Production Run of Radeon VII Just 5,000 Pieces, Company Denies it

[efikkan]

The sources for these claims seems weak, it could easily be that Radeon VII will have a limited first batch.

But if it turns out that this card is a low volume production, then it really doesn't matter how good it is, because you can't get it…

 

[Kaotik]

Is MI60/50 in that high demand?

Well, apparently they had sold over $20 million worth of them before they even started sampling the cards to customers

 

[InVasMani]

I feel like AMD needs to improve it's VSR support to make this card more compelling maybe then it's 16GB VRAM of HBM2 with twice the bandwidth of it's older Vega counter part would be more interesting compared to DSR at equivalent resolutions. As it sits now it's got the hardware, but software seems like a bit of a limitation on showing some of it's potential. In fact DSR is a reasonable talking point on why you might not buy this card as it sits which is kind of funny in it's own way.

 

[Captain_Tom]

Wow. That's quite an opinion. Best you keep it to yourself.

It's not an opinion. I am stating there are multiple videos where he spreads complete noob-like misinformation, and he acts completely disorderly when people (often) call him out.

1) He sets up (for example) a rendering benchmark and runs it on the CPU instead of the GPU's. He then concludes AMD and Nvidia render the same speed lol. Why would you mention this guy?

2) He proclaimed that "AMD is lying about prices" when demand overtook supply in late 2017. He doesn't understand basic market concepts. Why would you listen to this guy?

3) He has directly threatened other YouTubers on twitter with physical violence multiple times simply for disagreeing with him. You like this guy? You think he has good opinions?

Now why should I keep these facts to myself?

 

[lexluthermiester]

Now why should I keep these facts to myself?

Because your "facts" are anything but. All you are displaying is your complete ignorance to context. Do stop embarrassing yourself. It's more than just a bit sad.

 

[Xuper]

heh this guy JayzTwoCents repeated "I want to reserve my opinion on this" , obviously this guy was not happy.

 

[Captain_Tom]

Because your "facts" are anything but. All you are displaying is your complete ignorance to context. Do stop embarrassing yourself. It's more than just a bit sad.

Be specific about even one thing - what context am I missing?

You are quoting someone who threw a public temper-tantrum when he didn't get a threadripper sample (and literally said he isn't using it anymore because of his hurt feelings, suggesting "pay for play"), threatened someone he supposedly found lived close to him, and seemingly sent fans to threaten a couple of youtube tech people:

https://twitter.com/i/web/status/1032508074437763072

https://twitter.com/i/web/status/1026494411276840960

https://twitter.com/i/web/status/953880873954365440

Oh yeah, he can't even benchmark rendering correctly (and still hasn't corrected his video):

https://linustechtips.com/main/topi...dering-doesnt-matter-as-long-as-you-have-one/

^Jay saw the same render times between a bunch of cards and was confused. Good lord, rendering leverages the CPU mostly - you use GPU's for the previews. He doesn't even get how this works!


There, now I am providing links while you continue to spout complete nonsense, and try to shame my opinion.

 

[FordGT90Concept]

heh this guy JayzTwoCents repeated "I want to reserve my opinion on this" , obviously this guy was not happy.

Naw, that's what everyone should be saying at this point. Until card is in hand, we only have press numbers that AMD provided (hardly objective).

 

[Casecutter]

Vega needs bandwidth to get that performance.

But now with a 4096 bit wide-bus couldn't that that provide more bandwidth even using 8Gb HBM2? I would think to get another 15-18% over a Vega 64's 483.8 GB/s is not that difficult. Given Vega 7 works at as much as 1200 MHz (2400 MHz effective), which is more than the Vega 64 of only 945 MHz (1890 MHz effective). If they got a 1100Mhz clock (2200 MHz effective) would be more than enough to have it see an overall performance up-tick of 15%.

So using this Bandwidth Calculator I get 1126 GB/s, or about 10% off the Vega 7 1229 GB/s. I think that's enough...
http://gpubandwidthcalculator.totalh.net/?i=1

 

[lexluthermiester]

There, now I am providing links while you continue to spout complete nonsense, and try to shame my opinion.

See? There you go, you finally got something right. Now keep it to yourself because it's very clear context is lost on you and you're just bashing someone.

Naw, that's what everyone should be saying at this point. Until card is in hand, we only have press numbers that AMD provided (hardly objective).

Agreed and, like Jay, most people have said that. The general consensus is most people have cautious optimism.

 

[EarthDog]

But now with a 4096 bit wide-bus couldn't that that provide more bandwidth even using 8Gb HBM2? I would think to get another 15-18% over a Vega 64's 483.8 GB/s is not that difficult. Given Vega 7 works at as much as 1200 MHz (2400 MHz effective), which is more than the Vega 64 of only 945 MHz (1890 MHz effective). If they got a 1100Mhz clock (2200 MHz effective) would be more than enough to have it see an overall performance up-tick of 15%.

So using this Bandwidth Calculator I get 1126 GB/s, or about 10% off the Vega 7 1229 GB/s. I think that's enough...
http://gpubandwidthcalculator.totalh.net/?i=1

Whats the point though? It yields little improvement... Akin to only needing water from your sink and that is a firehose....

I honestly wish they would ditch HBM already as it doesn't seem to be helping much in the consumer/gaming realm. Its not like it was when first released and showed great gains at higher res, and especially 4K UHD gaming. GDDR5X and GDDR6 can work just fine up there. Is it really cheaper? How much power does it really save (considering these cards use a lot more power per watt)? I just think HBM was stillborn.

 

[Xuper]

Naw, that's what everyone should be saying at this point. Until card is in hand, we only have press numbers that AMD provided (hardly objective).

It's clear on youtube. He was not happy with Radeon VII.Many people like him and other were expected good performance at low price.this is what happen when People fall into rumor/Hype.

 

[Xzibit]

You are quoting someone who threw a public temper-tantrum when he didn't get a threadripper sample (and literally said he isn't using it anymore because of his hurt feelings, suggesting "pay for play"), threatened someone he supposedly found lived close to him, and seemingly sent fans to threaten a couple of youtube tech people:

Didn't he just announced he was going to have a fire sale. Get free stuff then sell for profit.

 

[efikkan]

Being a little optimistic is fine, provided it's based on realism.

But there is one thing that most people seem to have missed with the Radeon VII; AMD says it will be 25% better at the same power, and show benchmarks to indicate it's on par with RTX 2080, which would require an improvement of ~40% over Vega 64. Both of those can't be true at the same time. Radeon VII would have to be much better than 25% over Vega 64, or it's simply not on par with RTX 2080 in average gaming. I haven't seen any of the opinionators pick up on this. This is quite important as the card will be priced on par with RTX 2080.

 

[lexluthermiester]

Is that what this thread is going to turn into? A Jayz2Cents bash-fest? Quit being so pathetic and get back on topic.

It's clear on youtube. He was not happy with Radeon VII.Many people like him and other were expected good performance at low price.this is what happen when People fall into rumor/Hype.

No one has review samples yet, thus no one has a qualified perspective. Can't be unhappy with something you know nothing about. Did you even watch that video?

Didn't he just announced he was going to have a fire sale. Get free stuff then sell for profit.

He's had giveaway's for some of the stuff he has. Get your facts straight.

I honestly wish they would ditch HBM already as it doesn't seem to be helping much in the consumer/gaming realm. Its not like it was when first released and showed great gains at higher res, and especially 4K UHD gaming. GDDR5X and GDDR6 can work just fine up there. Is it really cheaper? How much power does it really save (considering these cards use a lot more power per watt)? I just think HBM was stillborn.

It would be interesting to see the difference between HBM2 and GDDR6 on Radeon7 GPU dies.

 

[Xzibit]

He's had giveaway's for some of the stuff he has. Get your facts straight.

I did, You should try it sometime

Also I desperately need to free up space so I will probably be holding a fire sale in the near future. EVERYTHING MUST GO!!! Well most things

No where does it say Giveaway.

 

[sneekypeet]

Not sure how fire sales, personal opinions on internet personalities, or much of what has been posted recently has to do with the topic.
Fair warning to all of those in a twist!

 

[moproblems99]

I honestly wish they would ditch HBM

I am pretty sure they are only using it to cut down on power draw. Doesn't it save them 50 - 75 watts over traditional VRAM?

But there is one thing that most people seem to have missed with the Radeon VII; AMD says it will be 25% better at the same power, and show benchmarks to indicate it's on par with RTX 2080, which would require an improvement of ~40% over Vega 64. Both of those can't be true at the same time. Radeon VII would have to be much better than 25% over Vega 64, or it's simply not on par with RTX 2080 in average gaming. I haven't seen any of the opinionators pick up on this. This is quite important as the card will be priced on par with RTX 2080.

Perhaps it's because the Vega 64 doesn't draw 300 watts and this one will? That extra 40 to 50 watts could provide that extra 15%.

 

[EarthDog]

I am pretty sure they are only using it to cut down on power draw. Doesn't it save them 50 - 75 watts over traditional VRAM?

sweet jebus... no... I dont think it's close to that much... I'm thinking maybe 20W?? No clue.

Lol, I couldn't imagine if it was 50W their cards would be using over 300W!

 

[Xuper]

Is that what this thread is going to turn into? A Jayz2Cents bash-fest? Quit being so pathetic and get back on topic.

No one has review samples yet, thus no one has a qualified perspective. Can't be unhappy with something you know nothing about. Did you even watch that video?

Simple.Price of Radeon VII.if it was $499 , I doubt he would pull those stuff.on YouTube You see some more his negative opinions.obviously he has the right to reserve his point.

It would be interesting to see the difference between HBM2 and GDDR6 on Radeon7 GPU dies.

this might answer

 

[moproblems99]

sweet jebus... no... I dont think it's close to that much... I'm thinking maybe 20W?? No clue.

Lol, I couldn't imagine if it was 50W their cards would be using over 300W!

So I am not sure how valuable GamerNexus is but this is what they say:

Spoiler: From GamerNexus

Speaking with Buildzoid, we know that Vega: Frontier Edition’s 16GB HBM2 pulls 20W max, using a DMM to determine this consumption. This ignores the voltage controller’s 3.3v draw, but we’re still at 20W memory, and no more than an additional 10W for the controller – that’s less than 30W for the entire memory system on Vega: Frontier Edition.
We also know that an RX 480 uses 40-50W for its 8GB, which is already a significant increase in power consumption per-GB over Vega: FE. The RX 480 also has a memory bandwidth of 256GB/s with 8GB GDDR5, versus Vega 64’s 484GB/s. The result is increased bandwidth, the same capacity, and lower power consumption, but at higher cost to build. In order for an RX 480 to hypothetically reach similar bandwidth, power consumption would increase significantly. Buildzoid calculates that a hypothetical 384-bit GDDR5 bus on Polaris architecture would push 60-75W, and an imaginary 512-bit bus would do 80-100W. For this reason alone, HBM2 saves AMD from high power budget that would otherwise be spent solely on memory. This comes down to architectural decisions made years ago by AMD, which are most readily solved for with HBM2, as HBM2 provides greater bandwidth per watt than GDDR5. HBM is effectively a necessity to make Vega at least somewhat power efficient while keeping the higher memory bandwidth. Imagine Vega 56, 64, or FE drawing an additional 70-100W – the world wouldn’t have it, and it’d be among the hottest cards since the GTX 480 or R9 290X.
https://www.gamersnexus.net/guides/3032-vega-56-cost-of-hbm2-and-necessity-to-use-it

Seems that 30W per 8GB ram could be reasonable, could be more...could be less.

 

[efikkan]

HBM might be more energy efficient, but GDDR is still easier to cool. So it comes down to what matters more; total power draw or power draw of the die.

 

[FordGT90Concept]

Does not compute: energy efficiency translates to cooler. Only reason why HBM is cooled with the GPU is because they're on the same package so they have to be. GDDR6 is cheaper (or will be), that's its only advantage.

AMD hasn't given any power numbers as far as I know other than saying it has 2x8-pin. That only tells us the maximum potential draw, not typical. Being 7 nm and more or less same transistor count, the power number should be significantly improved from Vega 64.

 

[TheGuruStud]

Whats the point though? It yields little improvement... Akin to only needing water from your sink and that is a firehose....

I honestly wish they would ditch HBM already as it doesn't seem to be helping much in the consumer/gaming realm. Its not like it was when first released and showed great gains at higher res, and especially 4K UHD gaming. GDDR5X and GDDR6 can work just fine up there. Is it really cheaper? How much power does it really save (considering these cards use a lot more power per watt)? I just think HBM was stillborn.

Vega is bw limited. That's why everyone wants samsung chips. You can hit up to 1.15 ghz giving good gains. More stacks would be used if it weren't so expensive. And if all the mauf hadn't lied about freqs and production...

Take a gander at benches of a healthily OCed vega. Nvidia becomes a joke as their minimums are awful (not that the ram is the culprit, but to give you the idea of gains on vega). Scoring way ahead in avg means nothing when your mins are under 60 lol. You'll need adaptive sync with that oscillation.

Production is the only problem. It saves space/power, so it's 100% the future unless someone wants to fund another type.

 

[efikkan]

How can Vega be bandwidth limited? It has the same memory bandwidth as GTX 1080 Ti and more than RTX 2080. It still lacks tiled rendering, but that shouldn't make this much of a difference.