Nvidia Maxwell-based 900 series cards now going into legacy support

Page 7 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.

Atreidin

Senior member
Mar 31, 2011
464
27
86
Watch the multiple people who claimed silverforce was lying not apologize. It's not hard to fact check, but you guys are happy to be wrong.

Also the attempts at derailing the thread are pretty obvious. If you want to discuss the claims AMD has made this really isn't the place for it. Trying to deflect with "what-about"-isms isn't doing you any favors.

The fallacious argument of "AMD lies too so it's not bad when Nvidia does it repeatedly" is only going to come back over and over again because it's a mindless, easy to make, illogical defense of a corporation that some people have an emotional connection to.
 

boozzer

Golden Member
Jan 12, 2012
1,549
18
81
I had hoped the extremeness of the statement would help people realize it was a joke (that and the "thanks Obama" joke). I bought it in January, it's like new, and I'm sure I could get $900 for it (paid $1200), but I don't have a PC to use in the mean time.
ahha, sorry but you are from texas. the obama thing is actually 100% real mine was around 1800$ sold it for a little less than 1400$. when I saw that 1060m was going to beat 980m by 10% or so, I had to sell right away.
 

David_k

Member
Apr 25, 2016
70
1
41
There is a note that clearly shows it. Stop being a nitwit.

TBH, Showing "RX 480" in the TITLE of this specific slide, where the WHOLE specific page was talking about cut Polaris 10 RX470, and pointing to a little "2" there without presetting immediately the actual note text is IMO a direct attempt to deceive without actually doing anything wrong, sure most of the people would "assume" and maybe some sites would do the same, and misinform even more people that this slide talks about the 480, as the event was about the 480, the name was WRITTEN in the slide title, but sure, they wrote on some small text along side some small number the numbers actually related to the 470 and not the 480, they did their part. so all good, GJ AMD. I'm sure they learned from the best.
 

eRacer

Member
Jun 14, 2004
167
31
91
"Up to 2.8x Performance/Watt With AMD Technologies"
And the footnote said "Using Performance/Board power". Radeon 270X board power is 180W, but actual gaming power consumption is around 120W.

The footnote says RX 470 board power is 110W, but the official spec now is 120W typical board power.

It also says in the footnote "Tests results are not average and may vary."

The "2.8X" number seems pretty meaningless with the "up to" in front of it.
 

MagickMan

Diamond Member
Aug 11, 2008
7,537
3
76
I don't know about that. Look how Pascal does well in what were AMD favored titles like Mordor, Ashes, and SW Battlefront where a Maxwell card still falls behind. I think Pascal has some secret sauce to mitigate the console effect somewhat while Maxwell is still vulnerable. Time will tell, but when they admit that Maxwell won't get specific optimizations you are basically gambling on Nvidia's benevolence if you have a Maxwell card. Historically that is a bad position to be in.

Memory bottlenecks have nothing to do with NV's "benevolence", if you OC a Maxwell card's VRAM to 8GHz or more that issue goes away. Unfortunately, reviewers have chosen to ignore that.
 

Raduque

Lifer
Aug 22, 2004
13,141
138
106
ahha, sorry but you are from texas. the obama thing is actually 100% real mine was around 1800$ sold it for a little less than 1400$. when I saw that 1060m was going to beat 980m by 10% or so, I had to sell right away.

Hah, wow. $1800. It should've had dual 980 desktop chips for that price. $1200 was a hard pill for me to swallow, but I only did so because it was insurance money that bought it.

The Obama thing... Might be real for some people, but I am not serious at all.
 

amenx

Diamond Member
Dec 17, 2004
4,007
2,277
136
The fallacious argument of "AMD lies too so it's not bad when Nvidia does it repeatedly" is only going to come back over and over again because it's a mindless, easy to make, illogical defense of a corporation that some people have an emotional connection to.
I'll bet far more people here have an emotional aversion to Nvidia than a 'connection', and most of them do not own NV or are affected by the issues (other than on an emotional basis). And most of it due, if I may be so blunt, to the fact that Nvidia consistently have the fastest enthusiast cards which may cause some non-Nvidia owners to feel that AMD has little epeen value left. So they must compensate by piling on a company that has contributed to their hardware inferiority complex by bashing it on end (unless they one day own something Nvidia). Witnessed the same thing with Intel vs AMD CPUs years ago. Same emotional rise, and need to bash the performance leader for deflating their epeens. Again, sorry if I may be so blunt ^_^.
 

Rvenger

Elite Member <br> Super Moderator <br> Video Cards
Apr 6, 2004
6,283
5
81
Well, looking at the thread title I see Nvidia, not AMD RX 480. Let's stick to the topic or I will be handing out infractions.

-Rvenger
 

boozzer

Golden Member
Jan 12, 2012
1,549
18
81
Hah, wow. $1800. It should've had dual 980 desktop chips for that price. $1200 was a hard pill for me to swallow, but I only did so because it was insurance money that bought it.

The Obama thing... Might be real for some people, but I am not serious at all.
gaming as a hobby is actually super cheap in comparison to most other hobbies. 400$ per year is super cheap. if money is tight, always go for desktops. they last longer and some parts can be carry over to new builds. you may think 1800 is alot but when gaming laptops first came out around 2003, alienware, the first company to sell gaming laptops, cost about 2200$ for the base model. was a 10 pound monstrosity. a couple of my buddies golf as a hobby, it easily costs about 3-500$ per month, and that is excluding the money they spend buying clubs. also compare to people who buy season tickets for their favorite teams? the money we spend on pc hardware is minuscule.
 

poofyhairguy

Lifer
Nov 20, 2005
14,612
318
126
I'll bet far more people here have an emotional aversion to Nvidia than a 'connection', and most of them do not own NV or are affected by the issues (other than on an emotional basis). And most of it due, if I may be so blunt, to the fact that Nvidia consistently have the fastest enthusiast cards which may cause some non-Nvidia owners to feel that AMD has little epeen value left. So they must compensate by piling on a company that has contributed to their hardware inferiority complex by bashing it on end (unless they one day own something Nvidia). Witnessed the same thing with Intel vs AMD CPUs years ago. Same emotional rise, and need to bash the performance leader for deflating their epeens. Again, sorry if I may be so blunt ^_^.


Are you trying to say the only people upset are always AMD customers? You are so wrong.

I have a 750 ti I plan to try and sell this week after learning that it's legacy. I use it for gaming sometimes and a I don't want to be left holding the bag. This matters to me, it's not my primary card but it's my best one on a sit down at a desk desktop so I care about its gaming future.

When I owned a 970, and it was my best card, the thought of it losing driver optimizations terrified me as we have evidence the driver team worked around the weird memory situation. Once we knew everything it was clear to me that without those optimizations it would probably age worse than Kepler in some cases as games expect 4GB which means they have no reason to stick below 3.5GB. And sure enough the 970 is falling behind the 980 in Directx 12 games overall, all because of an API that Nvidia can't "fix" solely via a driver but happens to be the future of PC gaming. I wouldn't own a Fury for the same reason, as I think game developers are only going to optimize to general architectures and not specific cards.

It is my recent experiences with Nvidia on which I base my opinion. I am not writing them off but they are no longer the default like before (when it took a great deal to sway me from team Green).
 

David_k

Member
Apr 25, 2016
70
1
41
The 980 pulls ahead vs the 970 @DX12 because its a stronger card with more hardware to utilize, dx12 is a more parallel API, so having more cores help spreading the load. nothing to do with nvidia's optimization, they optimized the cards to be as fast as possible, be sure that they didn't optimize the 970 more then the 980 so it "won't pull ahead". as a company they would want each card to be as fast as possible for its segment. faster than the competition if possible.
 
Last edited:

amenx

Diamond Member
Dec 17, 2004
4,007
2,277
136
Are you trying to say the only people upset are always AMD customers? You are so wrong.
If you carefully read the first sentence of my post you should see that is not what I am saying.

I have a 750 ti I plan to try and sell this week after learning that it's legacy. I use it for gaming sometimes and a I don't want to be left holding the bag. This matters to me, it's not my primary card but it's my best one on a sit down at a desk desktop so I care about its gaming future.

When I owned a 970, and it was my best card, the thought of it losing driver optimizations terrified me as we have evidence the driver team worked around the weird memory situation. Once we knew everything it was clear to me that without those optimizations it would probably age worse than Kepler in some cases as games expect 4GB which means they have no reason to stick below 3.5GB. And sure enough the 970 is falling behind the 980 in Directx 12 games overall, all because of an API that Nvidia can't "fix" solely via a driver but happens to be the future of PC gaming. I wouldn't own a Fury for the same reason, as I think game developers are only going to optimize to general architectures and not specific cards.

It is my recent experiences with Nvidia on which I base my opinion. I am not writing them off but they are no longer the default like before (when it took a great deal to sway me from team Green).
This is where I find a legitimate gripe for older Nvidia cards, and yes, incl possibly Maxwell. The 970 is in its own little bubble that needs specific driver support for it with newer games. So for the 970, that concern I believe is real. But the main arguments in the thread are about Maxwell (as a whole) becoming 'legacy' and therefore little to no driver support, incl the 980 and 980ti, and therefore Maxwell = Kepler.

Re (pre v2) Maxwell (incl 750ti), kepler, I believe that Nvidia has reached the limits of what they can do on the driver front. Their older arches are simply not capable of keeping up with new gen console developed games that are more suited to GCN and compute heavy workloads. It was a strategic lack of foresight in arch design (esp kepler). None of that seems to be argued here unfortunately (or I missed it). The gist of what seems to be argued here is that since Maxwell is now "legacy" (no longer being manufactured), that it will follow the same precipitous decline as kepler. A point disingenuously pushed with passion by many (not all) non-Nvidia owners in a similar vein as the "4th Pascal design flaw" thread.
 
Last edited:

Cookie Monster

Diamond Member
May 7, 2005
5,161
32
86
Re (pre v2) Maxwell (incl 750ti), kepler, I believe that Nvidia has reached the limits of what they can do on the driver front. Their older arches are simply not capable of keeping up with new gen console developed games that are more suited to GCN and compute heavy workloads. It was a strategic lack of foresight in arch design (esp kepler). None of that seems to be argued here unfortunately (or I missed it). The gist of what seems to be argued here is that since Maxwell is now "legacy" (no longer being manufactured), that it will follow the same precipitous decline as kepler. A point disingenuously pushed with passion by many (not all) non-Nvidia owners in a similar vein as the "4th Pascal design flaw" thread.

I think its a two way street. On one hand, if you design a GPU that will last several generations you potential risk sales for your next generation of products. Consumers benefit (and there is no incentive or reason to upgrade), but the IHV might be set back. Set back because there is less incentive to go buy the new generation especially when the cost of investment is increasing at a rapid rate vs the return in performance/efficiency improvements (unlike a decade ago).

Then we see the a more of a staggered release with features/architectures that suit for that specific time. Kepler performed well in games from 12~14. Maxwell took over from 14~16 and now we have Pascal. It also creates the need to upgrade. From a business point of view, its the one that will generate the most money so naturally that is what we are seeing from nVIDIA. I think this is also working favorably because the competition is lacking. That is why I suspect that by the time DX12 is in full swing (>2017), Volta could very well close the gap between Pascal and GCN even further to outright surpassing them even at their own game e.g. Async compute performance.

The company has a history of making record quarters after another one. Marketing can only get you so far because the products need to perform. And they have, at the right time and right place. So to me, it doesn't look like a lack of foresight but the opposite.

If they really wanted to kill AMD, they could have priced the 1070 at RX480 prices. It would have been game over especially when they have the similar power draws.

On a side note, how many users with 970s ever came across the memory segmentation problem?
 

OatisCampbell

Senior member
Jun 26, 2013
302
83
101
You know you have a weak argument when you have to dig back that far. We're talking about lies mostly about hardware that is currently in people's machines and available to purchase new now.

:\

I went back that far because it was a couple more instances of ATi/AMD issues that affected me as an ATi/AMD owner and fan.

I could probably do the same with NVIDIA parts, my point is no one has an argument where the company they prefer hasn't promised something and then not delivered.

Heck, when Mantle launched I thought it was "game over" for NVIDIA and that I would be in some new golden era of 3DFX like glory with my 290. Imagine my surprise when it basically got scrapped.

I didn't say,"D*** AMD! Liars!" though, I just figured the gains didn't turn out what they expected and they should put their money elsewhere. Kind of like their TruForm back in the day.

Someone should start a poll. I doubt there's 3 guys using a GTX480 on this forum, and I bet they don 't care about DX12. (if they DID, they sure as heck wouldn't be rocking a Thermi)
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Last edited by a moderator:
Feb 19, 2009
10,457
10
76
There's a difference from "Up to 2.8x with AMD technologies" and a footnote which goes to the RX 470 vs 270X (misleading marketing)...

Than outright lies. PR does this all the time. But we don't expect them to actually be so blatant and lie about specs or feature support.

Moving on, I do expect Maxwell to fall in relative performance vs Pascal & GCN in the next 6 months as more new titles are released. Especially DX12 ones. People will accuse NV of not optimizing it, but the real cause is the lack of proper preemption/context switching in the architecture. As games get compute heavy, Maxwell will tank relative to Pascal. Period.




Rvenger warned you all above to stay on topic or infractions will be levied.


esquared
Anandtech Forum Director
 
Last edited by a moderator:

David_k

Member
Apr 25, 2016
70
1
41
Yep, its "Legit", "Legal" but seriously on the verge of false advertisment.. "RX 480 Built on 14NM FINFET, OPTIMIZED BY AMD, UP TO 2.8X Performance/Watt with amd tech, Oh and BTW LOOK AT THIS LITTLE NUMBER 2 NO PERSON WOULD NOTICE ON THEIR SCREEN LOOKING AT DECENT DISTANCE"

I Hate shit like that, I dislike some of Nvidia actions lately and to some extant regret buying my 970 after 3 AMD cards I bought and were happy with, but still, the lack of proper competition from AMD vs NVIDIA and INTEL is really getting crazy. where on the CPU front you have $1700 part that supposed to be priced at $999, and a 1080Ti that would probably be close to the $999 Mark, AMD being competitive again would be REALLY nice, and it cant come soon enough. If you remember the 980Ti was priced according to the Fury X, without it who knows how hight would they price it. $700? $750?

And yes, Nvidia and Intel Put some effort at getting AMD where it is, the CPU era where even with inferior parts on CPU parts Intel Greatly outsold AMD, or the gameworks and closed source stuff (thought most of the gameworks titles do run fine on AMD these days).
Seems like AMD marketing team just can't do a decent job at building brand recognition so AMD could really shine, out side of forums where maybe 1% of the average joe comes to get advice regarding pc parts, and maybe at best 50% of them do actually buy AMD hardware. that's just not enough.


cursing is not allowed in the technical forums
Markfw900
 
Last edited by a moderator:

SolMiester

Diamond Member
Dec 19, 2004
5,331
17
76
In Doom I've seen benchmarks where Pitcairn beats the 780 ti. And before someone blames the 3GB, it's not bothering the 7970, which is beating the Titan by >10%. Kepler optimizations? Really?


Or in other words, AMDs driver support has been that bad for so long, they left 20+% performance on the floor.
 

dark zero

Platinum Member
Jun 2, 2015
2,655
138
106
Or in other words, AMDs driver support has been that bad for so long, they left 20+% performance on the floor.
I feel that nVIDIA Drivers was always giving ALL the performance on the card while the competition was 20% less and after some time, those performance is getting near and near.

Also nVIDIA is sadly known to have bad old drivers.
 

wahdangun

Golden Member
Feb 3, 2011
1,007
148
106
If you carefully read the first sentence of my post you should see that is not what I am saying.

This is where I find a legitimate gripe for older Nvidia cards, and yes, incl possibly Maxwell. The 970 is in its own little bubble that needs specific driver support for it with newer games. So for the 970, that concern I believe is real. But the main arguments in the thread are about Maxwell (as a whole) becoming 'legacy' and therefore little to no driver support, incl the 980 and 980ti, and therefore Maxwell = Kepler.

Re (pre v2) Maxwell (incl 750ti), kepler, I believe that Nvidia has reached the limits of what they can do on the driver front. Their older arches are simply not capable of keeping up with new gen console developed games that are more suited to GCN and compute heavy workloads. It was a strategic lack of foresight in arch design (esp kepler). None of that seems to be argued here unfortunately (or I missed it). The gist of what seems to be argued here is that since Maxwell is now "legacy" (no longer being manufactured), that it will follow the same precipitous decline as kepler. A point disingenuously pushed with passion by many (not all) non-Nvidia owners in a similar vein as the "4th Pascal design flaw" thread.


with maxwell fail really hard in DX 12, its really worried, and since pascal have better hardware in DX 12, what nvdia incentive to fix or improve dx 12 performance for maxwell user ?

hell look at the latest gamework title like mass effect 2, it ridiculously need 8 GB vram just to max everything.
 

dogen1

Senior member
Oct 14, 2014
739
40
91
hell look at the latest gamework title like mass effect 2, it ridiculously need 8 GB vram just to max everything.

that was mirror's edge and max settings are usually like that anyway. they usually just max out parameters with little regard to resource:visual impact ratio.
 
Status
Not open for further replies.
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |