Radeon RX 480 8GB reference review

You're almost trying to say that a 2500k/2600k isn't enough for a 480. Ok.

 

You do realize that if a user has bought a pre-built budget AMD rig in the last 4 years it is still running an older cheaper motherboard. This card is aimed at budget PC gamers but won't play friendly in some budget PC's. How many we don't know, yet.

Bottom line is this IS a problem as much as you want to sweep it under the rug.

 

i am saying.
you're coming from a 8800 in 2009... going to a 470 in 2011... then what until the 980Ti you had now, i was too lazy too look for more.

http://forums.guru3d.com/showpost.php?p=3044323&postcount=119

tell me how many mainboard changes did you made in that time. i counted 2 including the athl. 64 board, but i was again too lazy to make sure its not 3 or more.

i also actually found a review of the youtubers "shut-down" mainboard (release early 2009):
http://www.techradar.com/reviews/pc-mac/pc-components/motherboards/foxconn-a78ax-k-585968/review
so it isnt 2011 or 2010 (like others brought up first with no source) but early 2009 when that board came out.

so you would really support a person to expect that their 2009 motherboard is actually a good idea to pair with a freshly released 2016 tech GPU. in fast evolving end consumer market. and defend him that the card absolutely has to work?

now i am not super rich, which you can see from the dual 270x setup alone on which i sit for 3 years now, or at least dont think its worth paying 500-1000$ to have another performance than what my 270x cfx is giving me (smooth 80-100fps @ 1440p/SMAA/FXAA/TAA - high, generally speaking in majority of 2015/2016 tripleA titles), and thats enough. i know it's a special build and cost time to bring to its peak. but yes i don't go from 780Ti to 980Ti to 1080Ti every 2-3 years and pay up to 1000$ for that.
but i changed mainboard, GPU and CPU accordingly when dependencies occur...

for example... switch from one CPU to another... new mainboard -> has only pcie instead of AGP... ok means i have to buy new GPU even if i dont want to.

or other way round. new gpu -> only with boards with pcie, not available for old cpu socket -> means have to switch CPU as well.

also in the last 8 years, my mainboard changed 2 times. not high priced. eachtime 60-90$. still enough to stay in timeline, get use of new mainboard features etc.

there are dependencies that occur and are to accept. EVEN INSIDE A "SPEC".
and e"spec"ially OVER so much time we're talking about.
how can you not in the littlest bit accept this well known fact of persons who lived through several generations of hardware in their lives already, like you at least 4 gens...

but you want to say that a person who has a mainboard like the foxconn (from the youtuber) from 2009, and maybe then a gpu from that time (geforce 9x00 GT(X) ) will go from geforce pre-GT 1xx, GT 2xx... directly to RX 480 without questioning if his other parts are maybe incompatible or way too old for that GPU even if it has the same "SPEC"

over a time period of 8 years?

he can get a motherboard from 2013 for 20 BUCKS (LOL!?!?! how strange that even already sounds, who would these persons be i havent found them... people gaming for 8 years and then ... BOOM they saw RX480... now its time to upgrade...suuuure... pretty realistic) if he needs it.
and thats still the biggest of ifs.
so no, those people dont exist in the gamer area. they all have modern mainboards.

else you telling me they are playing 2015/2014 AA titles for which they would need a RX480??? with their 2009 rigs?

thats my "OK." so far.

 

This Polaris chip design is a failure for a 14nm die. And now the ref card turns out to be a turd as it doesn't do proper power draw balancing next to the inadequate heat sink slapped on it.
If it is a design fault on the card then reducing clocks and or voltage is about the only thing that could solve this.

This Polaris chip will have a very short time in the market. A refresh is already more or less confirmed.

Amd is in a hole with the R&D on this one.

 

(and this now is not directed to you eclap, but to all users here, just everyone):
i made a list of some GENERAL (not any card specific) wrong knock-out arguments people here make and other uninformed people jump on so often as well:

- "synthetic benchmarks like 3dmark score means nothing compared to real life gaming"
-> wrong. considering the newest version is used (dx11 features) it always represents a comparable score that shows a certain performance level that can be expected when playing a game that doesnt reduce performance with artificial proprietary features OR badly coded engine in general.
so. if someone has a 3dmark score of 10.000 it is unlikely his games will perform sub 80FPS in 1080P with high-ultra settings.
if someone has a 3dmark score of 3000 it is very sure that his games will perforom sub 80FPS in 1080P with high-ultra settings.
and that can be compared with other users scores.
so yes, relative performance differences are still comparable.
in terms of what we call synthetic in benchmarks, games are exactly as snynthetic as well. synthetic programs that generate drawcalls to be rendered by GPU and displayed.
if i say to a friend who has a different card, "open DOOM, put the ingame settings at exactly my values. and dont move after level x of game has loaded. whats your FPS?"
then i can compare...sure not bit-by-bit accurate drawcalls etc.
but.. yeah we can compare. its not like one of us is running 1080 and the other 1440p with the new "Hyper" settings and we seriously would compare that and think its comparable.

second
- "SLI / Crossfire works only for very few games"
its just wrong. i wont actually lose more words about it, because the sources where to get the info if seriously interested in true answer is out there.
and its not limited to this link
http://amdcrossfire.wikia.com/wiki/Crossfire_Game_Compatibility_List
but it's in a big enough numbers of reviews starting very early in gaming history with the older setups and also the newer.
as long as crossfire/sli has worked once... then the profile exists. and you can get it to work, if you want to (if you have a mgpu setup for starters)
and yes. cfx / sli configs/settings are badly documented. VERY BADLY.
i made some quick and dirty tuts for people with Fallout 4 or Witcher 3 problems concerning crossfire here some months ago already, and refreshed a very short while ago since there seemed problems occuring still.

- "this source is ****" or "kyle from [H] is not reliable" or "adoredTV is not reliable" or "techpowerup is not reliable" or "wccftech is also called rumorTV"
i am really very tired to talk about this. but it is the easiest way to evade a discussion. and when you actually measure the times that i said that in my whole forum life that a techsource is not reliable then you will find zero. and when we look at some other people it seems to be every second word out of their mouth.
sources are as reliable as far as you can understand and interpret the data presented, and eventually use the part thats correct in a "wrong" source an show up a mistake of a source to correct it.
i for example can and have today/yesterday repeatedly (i wont repeat here, thats your research work) corrected a techpowerup implication that fallout 4 wouldnt scale in SLI or crossfire.
people use the "this source is ****" in 99% of the time when they dont want to accept facts or continue talking about a topic.
if they dont want to continue talking about a topic, then i most of the time ask myself. if you dont want to talk about it, why are you in a forum, and why do you instead only flame/bash.

- "amd/nvidia drivers are generally worse than from the other company so i dont even think of buying the card from the other company"
yeah sure... thats why it works for almost (!!) everyone else.
also known as: individual problems being generalized
no further comment

- "look at your join date and look at mine"
http://forums.guru3d.com/showpost.php?p=5291649&postcount=32
instead of "hi welcome to the forums, why do you think polaris is an architecture to shake his head over?"
no further comment

i'll stop here, but there are so many disrespectful things. mainly against the companies, but in the end also against single users or user groups who have maybe on first sight exotic opinions, but when they want to prove it. the posts are not really read (just like this one of course also because of its length) and they are then called "CRAZY AMD FANBOY" or "CRAZY NVIDIA FANBOY"

now. i'll beg you to use the search function and just have a look over the history of the last 6 months. then you can see the people who are bashing, condescending, arrogant. regardless of if they are nvidia users or amd users. it comes down to a bunch of users. doesnt matter which card they have in the profile.

time for lunch.

 

Holy s***!

I could've had breakfast, lunch and dinner after writing that.

 

some ppl are WAY more efficient than the GPUs they promote!

 

well. i feed on coke an frito-lays (TM)

edit: and yes sammarbella i know it could be seen as wasted time. not that i was a regular but i liked this board a longer time ago when i had an older account. and it just changed.
i come from where vbulletin and phpboards were places that were free of trolls. at least 99% free of them.
and free of hate speech like what i encounter here.
and free of user bullying.

 

Wow, for a card that was supposed to take the mainstream by storm for a great low price of $200.
What do you think these 'mainstream users' have in their computers?

 

Running default has not caused any issue, other that out of spec test numbers. Also, the really bad out of spec numbers came from OC'ing and 4k tests. So if you are on an old motherboard and buy a reference card and expect to run it at max OC and then run into issues, well then....:cry:

Also, no one has proven that a card destroyed their system. No legit sites have been able to or are willing to try to destroy a system. Let's see some real damage and then let's talk about it......

Yeah AMD got cocky with their voltage....but this has been so overblown it is ridiculous. It is a ****ing video card....people act like Jesus just killed a litter of kittens.

 

I like it how you're now down to 2009. It was 2011 yesterday, then 2010, now 2009. Next you'll be telling us than nobody with a Celeron from 1998 should run a 480. Gotcha.

I've had 1 board since I upgraded from my q6600. It was an Asus z68 Pro and I would have it still, but it died. I've had my current board for some 3 years now, if not more and will most likely stick it out a bit longer.

 

I've had a z77 die on me 2 times and one z68 board. In a span of 3 years. My Q6600 had more luck it had 2 boards die in a span of 6 years.

 

are you too lazy to open a link and compare a motherboard model from a video with a release date?

someone stated 2011 at some point then a little later i went to the foxconn home page and saw the drivers that are there for that model have the date 2010, so i adjusted that in my future posts about it.

to finally find out that apart from the last driver downloads only being files dated from 2010 (and dont even contain windows 8 or windows 10 drivers (which probably means older drivers are not online anymore, or have never been online)
i looked up for the exact release date which can be found in the link that is in my first of the two very long posts on this page. beginning of 2009.

so a guy uses a oneslot pcie 2.0 spec board from beginning of 2009 to show that the card powers down on that one (IF ACTUALLY TRUE, but i believe him, because i am not a person who denies every negative fact about a brand i like)

and he says he wanted to build a what? a budget pc?
ok. so we build a budget pc and buy a radeon RX 480...
turns out radeon RX480 is only in the "budget" bracket because it has an incredible price.
i mean dude, theoretically. if nvidia tomorrow releases a card as fast as a 1080 with the same specs, then they call it 1080 II and then they say: "we sell this for 50$ each because we recently had a very lucky deal and special production circumstances evolving!"

then you would call this card "budget" for its price tag? implying that it is to be used with old hardware, because it "works best with that" and yes it is a budget card.

if you call a mid-high end card "budget" and "to be mainly used with older hardware" because of a price tag, well.
then you probably dont know that usb 3 sticks are also budget. but dont work with a 2009 PC despite being less complicated than a graphics card.

 

I will never consider other ppl opinions "wasted time" if they are not completely blinded by fanboyism (from any side)or they are not a pile of name calling and flaming statements.

They could be more or less in sync with my own point of view or even completely contrary but i want to read and discuss with them in here.

Black and white consensus is so boring for a discussion forum, i prefer colors.

Trolls are hopefully a mods business not ours...because if it was ours a very few ppl will remain in here.

Objectivity is a supposed gift mods have to properly "fix" incorrect posts, ban trolls and let me reuse another post, power efficiency you know...: cut Hydras heads... :chainsaw:

Please don't stop posting in here because other ppl opinions or desires.

 

All that is fine, except for one thing you haven't addressed yet and that's the fact he tested GTX980ti which worked fine. Power-supply was also upgraded from 400watts to 700watts and still no-go with RX 480.

 

Why only on this review RX 480 power consumption is higher than GTX 970?
Every other review, at least on well known websites it is vice versa.

You are trying to calculate peak TDP, but more important is total system wattage when gaming. There is a difference of 60-70Watts in RX 480 favor.
Your TDP measurement is misleading. Or maybe this is the only way to favor geforce cards.

 

get your facts straight bwoi
5 reviews in one place. 3 have 480 using more power than 970:

http://www.3dcenter.org/artikel/lau...-480/launch-analyse-amd-radeon-rx-480-seite-3

 

These are bold words.

The first phrase alone could be understood as a possible mistake or confusion in the measurement methodology...

...but after you wrote the second phrase it can only be understood as a whole as a clear accusation of partiality.

Good luck making "friends" in here, specially with the forum and site owner.