Radeon 5770 Outperfoming GeForce GTS 450 By A Lot

the cheap and powerful oldy radeon 5770 outperforming GeForce GTS 450, on these benchmark results (direct x 11)
http://kotaku.com/5636540/nvidia-geforce-gts-450-geforce-gts-450-sli-and-palit-gts-450-sonic-platinum-review

for those who bought GeForce GTS 450 for ff14, by having the weird idea that is the best card on the market

If you want performance, buy GTX. Buying GTS, especially a "mid-range" numbered model is not smart if you are expecting performance.

http://www.anandtech.com/show/3909/nvidias-geforce-gts-450-pushing-fermi-in-to-the-mainstream

gtx is not doing good on those benchmarks tho ati radeon stills does better (as seen on benchamrks ), i kinda knew it already, because the specs on ati cards have way more stream processors and higher core/shader/memory clocks for the same or better price than most nvidias.

Interesting lol, given that the game is supposedly optimized for intel and nvidia.

Optimized for Intel and Nvidia is a placebo sticker they place on any product. Also, I don't remember seeing any info about being "optimized for intel".

:)

Here are some other benchmarks that show Nvidia>ATI.

http://www.pcgameshardware.com/aid,743498/Geforce-GTX-480-and-GTX-470-reviewed-Fermi-performance-benchmarks/Reviews/?page=16

Are you trying to prove that a cheap midrange nvidia does worse than a cheap midrange ati?

Here's another benchmark that shows the 5770 #22 from the top.

http://www.videocardbenchmark.net/high_end_gpus.html

those benchmarks lacks of credibility specially with those weird list

Yes I can see how a third party benchmark software not affiliated with ati nor nvidia lacks credibility.

OH SNAP!

If you even looked at the list you'd see the GTS 250 scores like 500 points below the 5770 (confirming your post title). Both of which aren't even relevant cards for gaming PCs in 2010.

Generally over the years I have noticed a correlation between ATI having smoother framerate while Nvidia having better graphic quality. However, thanks to SLI and Crossfire, performance is just a second card away unless you end up bottlenecking with a slower CPU.

Also, I've been shied away from ATI due to horrific faulty drivers made many years back. Been happy with Nvidia for about 10 years now.

How about the test system? Is it a constant? The temperature of the testing environment the same? Are these factory defined clocks or manufacturer defined clocks? Why is there only 1 score, when there are multiple aspects of a card that can be tested. For that matter, why is it only 1 test ran when there are multiple out there. :)

Edit:
In response to the relevancy of a 5770 in a gaming PC, I don't know about you, but I don't have $300~400 to shell on a video card that is much better. My 5770 is a great card with amazing performance for such a mid range card. It is completely relevant.

http://www.videocardbenchmark.net/gpu_test_info.html

Says internal and users submissions all of which have been averaged. There's possibility for error sure. But even with averages, there are clear differences in performance given the cards listed.

With two 5970's I'm bottlenecking on CPU, and sadly an i7 980x is the best you can really get right now. Annoying.

XIV looks stunning on highest settings of course, but with the atrocious server lag on movement, it ends up looking like I'm dropping frames anyway. ><

They keep saying that the game will run great on Corei7

I'v been switching back and forth between ATI and Nvidia for the last 15 years, each has pros and cons, the holy war threads are getting old, anyway I think a well cooled GTX 450 with liquid cooling blocks will perform better, you will also need a crazy PSU.

As for FFXIV, Can't Blame ATI nor Nvidia for a poorly programmed graphics, wait for the official game to be released before you build your new PC. My current build is:

CPU: i7 Core 920 2.6 (overclocked to 4.2)
Mobo: Asus Rampage ii extreme with Sata 6g add on
Ram: 6g Corsair 12800 8-8-8-24
GPU: Radeon HD Sapphire 5850 1G (Overclocked 760/1170)
SSD: Crucial C300 64G (OS & FFXIV)
Storage: 2x 1terabyte 7200 WD
PSU: Corsair 750-Watt TX Series

With that being said, I had the game running at 1920*1080 with everything set at 2xMSSA highest all settings (except for Ambient Occlusion/Depth of field). At stock settings CPU/GPU I had few hiccups while turning the camera around fast, I had to overclock many times to get a stable prime 95 with good temps <80~ , and even then still having few glitches with camera speed turns. Although I'm really satisfied with the over all performance in camp areas, but that doesn't mean that we have to coop with a game being inefficiently heavily CPU/GPU/Memory dependent at the same time, After all the extra heat from OC is shortening our gear life span. So I hope SE do something about that.

Not to be too negative, but I don't get the folks arguing that the release client will be significantly different from the beta client. That doesn't happen in this industry.

Whatever beta client we have a day before launch will be the client on launch day. There is no conceivable reason to ever branch your source and have one client for beta and one for release. You just can't ship a product that way -- the logistics in the back-end code don't work.

It is conceivable the client will be patched to be better in the months/years after launch, but there will be no magical patch provided on Day 0 that wasn't there on Day -1.

I don't doubt that the tests show difference, but without a pure constant you'll get too many varying results, and it becomes less credible then say 1 box with swapping cards with the same temperatured environment.

One of the main reasons I say this is that the new Fermi cards are amazingly hot as far as I've heard and most people will find better cooling solutions for them compared to an ati card, which skews the natural benchmark in favor of Nvidia due to the better card getting better cooling solution where as if it was a constant you'd see performance drops on an Nvidia card at higher end tests due to heat.

Prime example of my whole argument.

Although this means squat in real life, it's just for the sake of consistency and credibility.

This websiteThe GTX 480 is hot. I know many enthusiasts are not going to be satisfied with a video card that idles at 73C (and even hotter in most SLI setups) and runs 93C under pretty much any kind of load.

My entire view about FFXIV (performance and hardware compatibility) condensed into one, clear, concise post.

Thanks Jaerik for saying what I couldn't express.

I've been having pretty decent luck with the stability and playability of the beta so far. Thanks to Cliff and his 11th hour mouse fix, the game is really enjoyable and looks amazing.

I hear people complaining about "poorly written graphic code" alot... and I just don't see it. The game looks amazing, and runs well on many different systems. Looks great and runs great on my crappy integrated AMD/ATI laptop on medium. On the desktop (Core i3 overclocked to 4Gigglehurts, 5870, and 4GB of ram) it runs amazing and it looks like the Crysis of MMOs.

I'm worried about the release though... I still get "You defeat you." messages... I'm afraid of what the Collectors Edition people are going to get... and even the standard buyers like me :|


Edit: and LOL at overclocking shortening life spans of hardware... takes your 20 year lifespan of a CPU to 10 years... I can live with that -.-

Inb4 anyone is still running a Pentium Pro 233mhz.


Moar Edit: Also, the 450 seems to slot nicely inbetween the 5770 and 5750 cards performance wise. It is not yet competitive on price... but it dropped almost $20 on release day and further drops are expected. If they can get it to $125ish, it's a solid card.

Tee hee

Poor graphics? Yeah uh... I'm running this on a HD 3650 1G with an AMD 4000+ (2.11Mhz) and the graphics are beautiful. I just seem to get a little bad shading and slow-as-a-turtle-pooing menus. I do get lag during battle, but between 50 people telling me 50 different potential reasons to why that is, I'll leave that aspect alone until it is properly dealt with.

UPS states my 5770 should arrive at my door today. I'll let you know how big of a difference the switch makes. But to be honest, If I could grab about 10 more fps and drop the amazingly slow menu speed, the game would be set (in my own subjective point of view).

GTX 400 series are more powerful cards, if you ran any other game like CoD etc you'd notice it. Nvidia just has some horrible driver problems at the moment that they're "looking into". Which is funny since XIV was designed to work optimally with Nvidia.
I use Radeon 5770, my GTX didn't even let the game boot.

I realize I made a mistake with this post. Your thread title talks about the GTS 450, not the GTS 250. The 450 isn't listed on that benchmark page (I guess because it was just released?)

This review shows the GTS 450 does better and worse than the 5770 depending on the game or program. No the 5770 doesn't outperform the 450 GTS by a lot (according to this review).

http://benchmarkreviews.com/index.php?option=com_content&task=view&id=416&Itemid=72

The 450 doesn't get dragged naked and screaming across hot coals by the 5770... but it does get beaten pretty much across the board. The 450 does however, give the same treatment to the 5750. So it's pretty much smack dab in the middle.

And I'm kinda torn on whether or not the performance will improve with a driver update. Nvidia has pulled rabbits from *** before... but I wouldn't base my purchasing decision on the drivers boosting performance by a large margin.


Also, the AMD refresh is about a month away... with a redesigned uncore and shader architecture, it's looking to gain about 30% across the board. It's likely the new 6000 series cards will slot in at current 5000 series prices, or just slightly above. Those of you looking to upgrade may want to deal with slightly crappy graphics for another month before taking the leap on a new card.

Either way you go, enjoy it :D

New drivers for the 400 series was promised to be out by this upcomming Wednesday.

Running on high/highest, standard shadow quality and no AA/AO and I'm experiencing little to no lag.

specs
CPU: Intel Core i7 860 @ 2.8 GHz
Memory: 8 GB DDR3
Graphics Card: ATI Radeon HD 5770 >_>

I mean, I'm not running on amazing resolution, but I'm also not on a widescreen monitor (I know, failure) so it's not really an option for me. >:

So 10 minutes through using the 5770 and I'm like "eh". Of course things are faster, but the whole snyth process is horribly lagged. OF COURSE (I know I know) I'll be told to wait until it goes live, but as Jaerik has previously mentioned, there won't be a big difference. I hear things like "debug will be off" and "last minute patches will fix bugs that are using way to much power", but if there is ANY server lag (as Jaerik is mentioning) when this game goes live, I will not play. That has to be one of the most unacceptable problems for any MMO. No amount of upgrades I can implement will fix a broken game. Well back to my point.

I would recommend getting something better than the 5770. Even menu scrolling is horribly slow. Or double up on them. I think that is what I will have to eventually submit to doing. But before I make any concrete decisions, I will be giving 14 a month to get smoother.

Edit: let me be specific about "menu scrolling". When I go to select items for synthing, the lag is mindblowing.