Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
...9 seconds for the Blender BMW benchmark scene.
Less than a minute and a half for the Junk Shop scene.
Yeh, I know.
BTW, from what I've been seeing, the card can't currently ( and perhaps never will be) utilitise the full 24GB due to it not being a Titan, despite what was said in the presentation by Nvidia.
What do you mean by that?
$1589 (USD) for the 3090 model I think I'd want at Newegg. That's pricey. 3080 same vendor line is $799. Is a 3090 twice as good as a 3080?
If you are doing anything requiring lots of VRAM, it's worth it if you get the $1499 card from Nvidia as it has more than twice the memory. The other card with 24 GB of VRAM is the older RTX Titan for $2499. So the 3090 is a bargain. If you never fill up your VRAM and drop to CPU you don't need it. If you do this just as a hobby and just still frames you don't need it. If you do animation it can be quite helpful because animation rendering tends to need more VRAM for most. It depends on what you do and what you're trying to solve. If you're mostly a gamer you have other choices..
If you need more than 10 gigs, it's a good deal. If you need more than 20, it's a great deal. If you need more than 24, it's the only game in town.
If ture, that would suck and this card would completely fail to justify its price.
But I can't imaging Nvidia came up with a 24G card without letting content creators to utilitize all the vRams, that decision would be unwise, plus no games in the foreseeable future will come close programmed to consume that amount. Titan is just a class name, it still use the same old gaming or studio geForce drive to get access to all 24G rams, so let's wait and see the new benchmarks on 3D apps.
Every ISP in the US that I'm aware of has one. People bumped up against it them when they started working at home. You may have a hard time finding what it is or that it exists at all, Comcast tried to hide it in Chicago where I'm at as it violated their contract with the city but my wife and I went through it in less than a week and suddenly our DL speed went to nothing. After some pointed discussions, and then "offering" to sell us more data, they wiaved the cap, turned out the city had told them to.
There are plenty of reviews that say that Nvidia has said this isn't a Titan and has certain Titan features disabled. Gamers Nexus pretty much confirmed that a Titan will come later.
I really would hold your cash till some more reviews come out. This whole launch has been fishy and this card has had me scratching my head from the very start.
...on the other hand, if you create epic quality scenes, are terrible at compositing/postwork, and want to avoind render jobs dumping to the CPU as much as possible, it still sounds worth the price.
I can even make a Titan-X cringe.
KK you need to get it!
For those of you lookiing for 3090 Octane and Blender benchmarks, as compared to 3080, 2080 Ti and other cards:
https://techgage.com/article/nvidia-geforce-rtx-3090-proviz-performance/
...that's oddly who I ended up going with for they offered the best deal for the speed I wanted. 15 - 25 MPS is pathetic as a 500MB Daz file would take forever to download (I've downloaded individual content zips in excess of 700 - 800MB from the store here) and any "hiccups" would mean a failed downloadand having to start all over again. 300 MBPS would be nice but just outside my budget. 1,000 MBPS isn't available and rather moot anyway as the building only has cable and "copper wire" (phone) from the street to the individual units. I'll have to check if Portland has a similar contract like Chicago, however, I don't work online from home all day (retired), and again don't stream online games (which can chew up cap space quickly) or films/television so I should be OK. We only have a few ISP carriers here (odd for a fiarly sizeable city) several of which are wireless only which I have no intrest in given previous experiences.
I find it strange that here in Portland, which is considered a very "techie" minded city, there are only limited areas where you can get high speed fibre optic, and, that it is not available in the neighbourhood I live in (which is a fairly upscale one).
Here's a screencap of the statement made by Nvidia to LinusTechTips for their review regarding professional workload optimizations on the 3090:
What this indicates to me is almost certainly no TCC support, since that is usually a much more AI training and large dataset analysis associated feature. Which imo makes the 3090 pretty much a no-go for professional workstation use. Including for rendering in Iray (despite its undoubtedly stellar rendering performance.) Having a 24GB framebuffer GPU is only useful for a very specific set of computing tasks (among them 3D rendering.) And paying a premium price (even the relatively low one of the 3090 vs. a typical Titan) for a GPU with that sort of hardware but a purposeful lack of software support for all but one or two of those specific tasks - even if one of those happens to be the one you primarily intend to use it for - simply isn't a feasible thing to be doing from a business perspective.
Chances are that if you are the type to be needing up to a 24GB framebuffer for 3D rendering, you are also most likely going to end up in situations in the near future where you need a 24GB framebuffer for eg. ai processing (because as folks at places like Nvidia love to tell us everything is going machine learning - see also: DLSS/ai denoising.) Or CAD work (because being a successful independent 3D artist in the current world means being able to do whatever kind of computing work happens to come your way.) So paying for that level of hardware on a card that will only support in software one or two of those things specifically is just handicapping your future professional growth.
ETA:
All this does make me increasingly interested to see what Nvidia's plans are for the Titan series gonig forward. I can only assume that the 3000 series Titan (assumnig it ever comes) will be a 48GB card (the next step up in VRAM capacity from Nvidia's existing pro card lineups) with a fully unlocked GA102 die - making it approxiamtely as much faster as the 3090 as the Titan RTX was versus the 2080 Ti. Most likely for around the same price as the existing Titan RTX. As to when to expect to see such a thing - I'd say around Christmas time. Because Nvidia needs time to get their die fabbing process refined enough to start turning out enough fully functional TU102s - and that is how long it took them to do it the last time with the Titan RTX.
Then I sympathize with you too, Comcast seems to be the worst of the lot. Glad that you got it sorted, one way or the other.
Chicago has 2 providers. Comcast and RCN, plus a few wireless ones but I'm not interested in wireless internet. They only offer gigabit fiber to residences in the highest density areas and I don't live there. So Portland isn't unique in that regard.
Comcast from what I hear isn't the worst ISP. I'm not saying its great just that I hear a lot worse from others in the tech field about Cox and Time Warner (who has changed names but I don't remember to what).
Wait, you're right... not Comcast but Cox.
But I have Time Warner, now Spectrum, and the only thing I don't like about them is the assymetric connection. I had to get 940 down just so that I could get 35 up instead of the usual 15. All my backups are off site and it was killing me. Meanwhile, friends on the other side of the city have symmetrical connections with Frontier, nee Verizon.
Back in NY when fiber was new, they sold to us as high speed unlimited, and every few days would throttle me down to like 50 kb/s because I was using "too much". I would call and cuss out a manager, threaten to switch and they would uncap me. That went on for months until they just stopped their crap. When I first moved up here, the ISP all had low data caps, but thankfully the provincial government made that illegal. Their meter was very innacurate, like my router data counter and their counter would be like 100's of gigs different, and I would get ridiculous overage charges. There is no reason to have data caps on fiber lol...
Truly. This whole thing is ludicrous. The US government gave these companies billions in tax breaks to upgrade their networks and they didn't do it. So when the pandemic happened and everyone started to work from home the residential lines couldn't take the strain they had to do stuff like get Netflix and youtube to throttle their content.
The truly galling thing is my brother lives in Chattanooga where the electric utility put fiber into every home and now the city is the ISP and you can have gigabit internet for less than $100 a month (which is roughly what I pay for 300 mb/s)
Titan had always been a hybrid class GPU for people like me who create 3D content and wish to play video games on the same machine, since quadro cost too much for indie & freelancers to sustain and can't play games for squad, that's how titan came to be Titan in the first place, the price is/was always in between quadro and consumer class GPU. The only useful feature on Titan vs it's lower peers in rendering? It's just the much larger vRam pool(superior time, speed and capacity) 3090 by any measure looks like a titan sound like a titan, will likely perform like one too, Nvidia would be very foolish to blockage all its vram access to content creators. As for gaming, none of previous gen titan had any substantial advantage on their ti peers sometimes even worse. We will see the benchmark scores, not many real person able to snatch one yet.
edit: just started reading new replies
Sweet Jesus, that's insane.
I just watched a couple other branchmarks on rendering apps as I type in these replies, averagely x2 faster than 2080ti and 20%-50% faster than titan, unbelievable, too bad I wasn't able to get one this morning.
Nice info, at this point after watched all recent 3090 benchmark on rendering apps and saw its performance in iRay/fstrom/octane, there's no double that 3090 far supreassed titan by good margin, as for titan's special feature on machine learning & data chrunching, well, I doubt any lab would use titan/3090, there's another class CPU called Tesla. Anyhow, 3090's release is a great news for 3D community, now only if Nvidia have any in stock.
also, Are you a Daz dev?
Watching this video is quite interesting, and talks about possible Ampere Quadro configurations.. One interesting part that is discussed is why the silliness of 10Gb of GDDR6X ram instead of 12 Gb of GDDR6 ram for the 3080.. He says that the 3090 should had the new GDDR6X and that the 3080 just have GDDR6, also it seems that the new Quadros will use GDDR6 too..
You definitely don't want to touch this if gaming is all you do on your PC when the 3080 exists.
Windows will reserve some of the 24GB of RAM; this can disabled in software on the Titan and Quadro cards I understand.
Yeh I watched that earlier; he seems to be continually spot-on with his information. Very reliable.
Foolish they may or may not be.
... Howerver, it currently doesn't have access to the same tools as a Titan. So it will reserve some RAM. There's a LinusTechTips vid that criticises Nvidia for doing that.
I wonder what kind of performance the new Quadro 6000 would get you on Iray. I took a look on the benchmark thread, but nobody seems to have benched it with more than an RTX Quadro 4000. And what's that thing going to cost anyway, $4000-5000?
Yup he often states that he has sources from within Nvidia, with that can only take his word for it.. But it seems that this time around the Quadro's are supposedly going to be alright for when one wants entertainment at the end of a hard days work..
What will be interesting is the cost of the base model since the flavours this time around are supposedly 16 Gb, 24 Gb and 48 Gb GDDR6..
My guess is that there will be a Titan sometime. I can't see Nvidia changing the 3090 so it is more compute (able to use all RAM mainly), despite LinusTechTip's plea. But he did get to check out the 8K gaming curtesy of Nvidia, so I wouldn't be surprised if the keep an eye on his vids.