Gadgetory

we have to break out this a 88x motherboard for its PCI M PCIe potential for video cards that's because we spent $50 on eight video cards via the vision tech mystery box so I don't think they knew who I was when we ordered it I don't I don't think they recognize the address Patrick and I both ordered so we sent some of the Patrick's house some some to me that way we could mix it up in case I tried to sneak something in there based on the ordering the destination what we end up with these eight before that this video is brought to you by the gigabyte z3 90 ARS master motherboard which comes equipped with one of the more powerful z3 90 V RMS for heavier overclocks on the new 9th gen Intel CPUs the ARS Master is also one of the few motherboards with a real heatsink this generation featuring a mix of high surface area fins and looks oriented cover blocks oh and it's also got updated RDV illumination learn more at the link below so these eight are the Radeon HD we've got a 64 50 in here we have a couple 54 fifties actually a two gigabyte note 512 megabyte we have a two two gigabyte 5450 sg4 8400 GS 512 megabyte a 2400 pro 256 megabyte and a Radeon X 1300 and 2600 XT so we paid about five to ten dollars per card depending on if it's PCIe PCI or AGP and then decided after we saw what what we received from the mystery bundle to see if they're any good tested on sort of semi modern tests vision deck has done these before we ran an article in 2014 when vision tech first did their mystery box GPUs vision Tech's been around long enough that they have probably a warehouse full of old inventory that they can't get rid of at this point this is a good idea to move through some of it now the usefulness of the cards to you probably not that high but if you just need a display out on an old board I guess really specific then maybe it's worth it either way though we wanted to test it out and originally in that 2014 article and we can show it on the screen here the cards at the time we're less they cost less than they did for this round of cards you know they say the same one so previously the AGP cards were a dollar the Radeon 2000 3000 series cards were $2.00 and then the three the 4000 5000 series cards were three dollars and for this round visiontek increased the price to five dollars for most of them and then ten for a quote high-end or once high-end anyway PCI Express card and what we ended up with is what we have on the table so these are not particularly high-end devices we had to roll back our testbench and methodology a number of years but we do have tests on them and then as noted earlier we did two orders so one went to me had my name on it the other one went to Patrick had his name on it and it was just to make sure that nothing unrealistic slipped into the boxes in case you were to order one we didn't want to say you'd end up with an Rx or r5 card or something like that as for the cards available the 8400 GS and HD 6450 are the only two new options that we received that were not offered in the older and cheaper boxes from before moreover at the time of working on the content for this all the mystery GPUs have disappeared from visiontext site once again so either we bought them all or visiontext waiting for that HD 5450 market to really spring up the 5450 is a crossfire capable GPU it's one of the more interesting ones we got so getting two of them meant that Patrick had a chance to compete with my 6450 that I got in an obviously identical cards or at least identical amounts of VRAM would be ideal but we did experiment with that briefly using the - 2 gigabyte cards together although it was against the sort of the spirit of the competition since one of them was mine and not Patrick's but he stole it from me anyway one of the two gigabyte 54 50s was a terrible overclocker so we continued with the mismatched pair instead although we're mostly scouring to see who got the better mystery box of GPUs we did have to work together on some of the tests so for one of them we used identical we used the 54 50s that had higher overclocking capabilities to just get a number on the chart that had some chance of doing any framerate that was reasonable and for that we use the higher overclocking 54 50s in fire the crossfire overclock used for testing was a 700 megahertz core clock up from about 650 and 720 5 megahertz memory up from 500 megahertz not too bad but that's about what was stable and that was on both cards so they could have been tuned individually but doing them together was a bit easier for this and frankly you're probably not going out to buy these anyway then finally the lower capacity 512 megabyte 5450 manage the memory clock of 900 megahertz without any visible artifacting and all this was done through Andy's control panel so the plan then visiontek may not have stacked the deck against one of us or the other but one of us did get a better deal and clearly highly scientific benchmarking is the only way to know for sure which selection of five-dollar decade-old GPUs is superior and with that in mind the first step was to figure out a compatible test bench testing PCIe GPUs is easy we have shelves of motherboards and all of them have PCIe slots dating back to the beginnings of what we were doing here we could plug in every single PCIe GPU that we've ever had and to still have slots left over PCI and AGP are a different story the only motherboard with PCI slots that we've used in normal benchmarks lately is the asus p8 z68 - b pro and that's the board we use for testing our 2500 k and 2600 K Intel CPUs this board wouldn't post with a PCI card connected unfortunately and neither would the nearly identical p8 P 67 deluxe that was sent into us by GN discord moderator Liquid Paper it's probably a solvable problem but it wasn't worth spending yet more time to figure it out so we began trying more boards that were sent into the P o box by viewers and we had a few options here the EVGA 790i FTW was promising but damage to the CPU socket meant that it would constantly reset BIOS settings preventing the selection of PCI cards from the primary display the gigabyte G am 7 5 0 sli - D s4 is an an - plus board and therefore needed an old Athlon 64 chip socketed to apply a BIOS update before it would support the Phenom 2 X 6 1090 T that we preferred for testing and it also required a CMOS battery replacement it's old it would boot and then using a drive with Windows 10 pre-installed but Windows 10 refused to even attempt a clean install this led to some increasingly desperate attempts to boot into Windows with the Nvidia card connected long enough just to try and figure out its name which didn't work the system worked fine using Windows 7 until we started swapping GPUs afterwards the bios issues reappeared so we had to move on from this one and it didn't support crossfire support anyway and could only fit four gigabytes of ddr2 because we don't have any sticks a larger than one gigabyte so after two or three days of not getting much done with these boards we gave up trying to be creative with the viewer donated boards and we dug out our 888 2x pro this Asus board right here and this one is something we use back when Andy was sending us lots and lots of their ap is to benchmark and review and I think we ended up using the 7890 K APU for this process for benchmarking and this board finally did actually work so it took a while but we found one that could run PCI cards without any issues finally as for the AGB cards we just didn't bother we had a board here from a viewer but just it by the time we got it anyway it didn't work anymore so the two cards that we got that were AGP will not be in these charts they are the Radeon X 1300 and the 2600 XT but realistically if you bought one of those for an actual use you're probably using it just for display out on some old system to get some files off or something like that so PCI and PCI use what we're testing for this we tried to keep the benchmarks semi-modern while staying within the hardware capabilities but the PCI cards are not dx11 capable and in fact had plenty of difficulty just launching Windows at least Windows 10 we used 3d marks cloud gate and sky diver benchmarks primarily with Cinebench r15 and the OpenGL test for extra testing Unigine heaven and for a game Metro last light tests were run at 720p because Patrick's 512 megabyte HD 50 or 50 didn't seem to like running 1080p at all at least not in crossfire and he wanted to be able to use crossfire as an advantage because he's a cheater rest assured the GPUs were still the bottleneck at that resolution the scores recorded are all the benchmark provided numbers which is not the method we normally use for scoring games like Metro last light but it was hard enough just getting these tests running at all without adding third-party login software to the mix and then there are plenty of additional challenges too just because it's all old hardware we run the Cinebench OpenGL test for extra verification during CPU testing although those numbers don't make it to the published reviews it's intended to be a GPU rather than a CPU test but r15 was released in 2013 and it take a pretty low end GPU these days to bottleneck it fortunately that's exactly what we have the HD 64 50 Bostock and overclocked handily tops the chart at 16 point 1 FPS average this is the only one of our tests that doesn't utilize crossfire so the HD 5450 is all scored roughly the same in all configurations but with slight advantages from overclocking that was led by the extra fast a 512 megabyte card the 8400 GS brought up the rear with an abysmal 3.5 3 FPS average but at least it actually ran the test which the HD 2400 Pro did not Cinebench runs a validation pass before the OpenGL test to make sure everything is working properly and the HD 2400 Pro repeatedly failed this validation pass so for the start on this one we've got a mix of cards in terms of whether Patrick's or mine performed better although there is some clear distinction and who got the better value Metro is the only real game on our list because it's one of the very few games that we've tested in recent memory that could actually run these cards the benchmark was Ronda in dx11 at 720p with high quality which ruled out the older non dx11 capable cards we ran the entire benchmark pass and took the provided numbers from the end rather than using our usual method of using frame logging software after all when the benchmark average is under 10 FPS the 0.1% lowers really don't seem that important at the top of the charge is the pair of overclocked crossfire 54 50s with an average frame rate of eleven point nine that might actually become playable with some settings lowered but impressively even the stock 54 50s in crossfire beat the single overclocked 64 50 although it gained a good bit of performance from overclocking the 512 megabyte 54 50 runs an impressive 5.4 FPS average which ties it with the 2 gigabyte HD 50 or 50 the straight in that memory isn't the limitation here but rather the core this is further illustrated a second time when our overclocked to the or 50s effectively achieved the same 7 FPS average within overclocking variants between the two crossfire scaling versus a single card wasn't impressive 83% stock to stock it's rare to find that kind of scaling in modern titles but still happens sometimes Cloud Gate is a testament for low-end systems like laptops but it works for ancient GPUs as well the two graphics tests are the relevant ones here as it would take a really awful GPO to bottleneck the physics test instead of our 7890 K only the HD 2400 Pro performed that poorly again but this was also the only graphics test that it could complete at all the 8400 GS consistently crashed after completing the first graphics test at about 1 frame per second but it made a valiant effort in the very least the card still can achieve 60 frames per minute which sounds impressive if you zoned out and missed the change metric of time there 2 minutes on the high end of the chart the overclocked sat crossfire cards dominate once again managing an almost semi bearable 22 point 3 FPS average in the first graphics test in 18.4 in the second the overclocked 64 50 scored almost as high as the stock crossfire 54 50s roughly in line with the scaling seen in the Metro test skydiver is a real test in the second to last one here in the sense that this one was made with desktops in mind rather than laptops and ap use there's also no pure physics test here instead making do with a combined test that was absolutely GPU limited with these cards anyway the gradients of scores looks a little smoother than in cloud strike at a glance anyway but that's only because the lowest end GPU is could run cloud strike whereas sky diver is beyond their capabilities and they were removed from the test as a result crossfire scaling is again about 2x in this benchmark with the worst card operated at 5.7 FPS average and functionally tied with an error margins to the HD 5450 512 megabyte card this again illustrates that the extra v round capacity really doesn't help in these benchmarks Unigine heaven is the last of the tests and it largely confirms what the others already showed the overclocked crossfire cards were the best scoring by far and the stock to gigabyte 5450 lags ever so slightly behind the stock 512 megabyte one the crossfire 54 50s ended up at 326 points when overclocked leading stock crossfire performance by 28% the stock 54 50s in crossfire led the fastest single card 5450 stock score of 133 by 92% posted nearly perfect scaling all right so now I'm joined by Patrick he did all the testing on this stuff and encouraged me to buy the cards so we spent 50 bucks on this Patrick what what's your conclusion I mean some of these you know the display outs are DVI and VGA and I think s-video yes so usefulness is limited but would you would you would you do this would you actually spend $5.00 well it's it's a tricky question because would I spend $5.00 on one of these if I just needed an HDMI out or just something to like slap into a test system really quick yes but their mystery cards so there's no guarantee of what you're gonna get it's probably gonna be in HD 5450 statistically speaking yes statistically speaking we got what three about four old three of them three and 16450 and we paid more for one of them on the other I think yeah so we paid five five dollars for one of the 54 52 gigabyte cards and 10 for the other one there so clearly at visiontext there's no actual I don't think they actually have bins of like this is five and this is $10 I think it's like here's a bin of garbage and just put any of them in the box and ship it it's probably true and they also took down the mystery box offers after we bought these so we may have just bought the whole bin of trash eyes on them I there have to be more I feel like there had more probably because they did this before like in 2014 they they did this people didn't buy enough room and they were the same cards then yeah but which which was the one that you liked for it's cooler so these are the a JP cards here we didn't test either of these I there's no reason to think that there working there probably working fine we just don't have any AGP boards around but this one is heavier which makes it better yeah and it has a picture of a lady on it and this one doesn't so I'm declaring this one the winner yes and there's at least 10 polygons in her face yes so this is clearly capable of very high resolution high poly rendering also it has a backplate on though which is better than some r-tx cards presently but oh and it's got their own pads on it too if we're just judging by a cool Aurora though I think this gives it a run for its money yeah that's great really good cool all right guys sitting at a desk very prophetic sign you know it's profession right I had forgotten yes so I don't know I think I think we're probably in the in the not worth it category for the most part if you don't know what you're gonna get it's not worth it because you could always end up with if you buy one of the PCI cards something like this which coincidentally happen to have a cable that fits this but it's not a normal DB il would even know what card this was for a long time yeah and yeah is that the Nvidia card yes that is the one the one Nvidia card which didn't have its name anywhere and you you just tried to boot into an OS to get it to tell us what it was yeah so fun though fun experiment for a couple bucks I guess if you're in the mood to throw a few dollars away and I don't know it's basically a loot crate IRL is what it is except then you can't like you can't wear it and fortnight I think that's what people do and crates and you can't resell it on eBay you can't resell it on eBay yeah yes oh well you could for definitely less than you paid for yes so probably not worth it fun experiment though if you see anything similar to this out there alert us to it you can post a comment below with any suggestions content ideas you have in this vein and we'd be happy to look into it because it's it's in the least affordable content so 50 bucks down the drain but you get to see what the mystery was and and it's now it'll probably back around Christmas again because that's what visiontek tends to do since I don't think they saw much of anything at this point other than the mystery stuff and maybe some all these Jesus speakers well it's true thank you for watching subscribe for more we'll see you all next time