Gadgetory

welcome back to harbor unboxed okay so a few weeks ago now we reviewed the geforce gtx 1650 for the first time and like most in the tech press i think it's fair to say we weren't overly impressed but that was okay because most of you weren't overly impressed either we just have a few Nvidia fans they've got a bit upset with our take on things but I did address those criticisms in a replying to comments episode the key argument was power consumption the GTX 1650 consumes quite a bit less power than the rx 570 and models without a 6-pin PCI power connector can be used in older om systems and this provides a cost-effective upgrade giving them a new lease on life I questioned just how many of these older systems could take full advantage of a gtx 1650 and I was seriously doubting just how cost-effective the upgrade really would be therefore I pledged to track down a popular early MPC that didn't have a 6-pin PCI II power connector and the power supply couldn't easily be upgraded to get around that one this led me to the HP elite desk 800 g1 so one of these things here it's a computer that most of the people who were in favor of the gtx 1650 saying that it wasn't a pointless product well this was the system that most of them recommended so I went ahead and got one of course being an OEM PC buying novice I purchased the small form-factor version first so the the lower profile version of this system you see here and because there are no low profile gtx 1650 models available right now for reasons unknown strange reasons they must be because yeah a bit bizarre but anyway because of that I couldn't use that system so I had to go and get the mid tower version in my defense though 99% of all HP elite desk 800 g1 selling on ebay combo you are actually the small form-factor versions however by paying a little bit extra I was able to acquire the mid tower version sporting a Core i5 4690k eiga bytes of ddr3 1600 memory and a 128 gigabyte SSD now the plan for this video was very simple purchase an HP elite desk 800 g1 kind of got that right second time around got that right then the plan was to stick in a gtx 1650 at 1050 TI and an RX 560 obviously not all at the same time but sure you guys knew that's not what I meant anyway then I was to benchmark about a dozen games and then compared those results with our high-end test system the testing I've already done and the specs for that system and this system will be in the video description and weight plan was to do that and then see how the margins changed did they come together for example so this I suspected would be a bit more CPU bound it's a core i5 with ddr3 memory so it's only a quad core and would that create a more CPU bound situation that would see the margin reduced between the 1650 and the 1050 Ti and well I suppose spoiler alert this is exactly what I found at least for the most part okay so I got my shiny old OAM PC a little over a week ago now I've been testing it pretty much for the whole week so it's very late right now as I'm filming but I just I have to wrap this up I've got other content to get on with and this has taken me a whole week so yeah if I look a bit tired I am sorry for that but I am quite tired anyway I got this old banger and I thought yippee I'll just start benchmarking that's always I've been interesting to look at a system like this and look at all the hardware and see how performs in modern titles anyway little did I know that it would end up being much more involved than what I expected and I'm not talking about having to update biases or anything like that the second it arrived I threw in the GTX 1650 it's currently in there now I stripped out the 128 gigabyte SSD and I replaced it with a 2 terabyte model so a bit of a quick upgrade there but that was really necessary because I had to put a fresh copy of Windows on it and then I had to copy all my game files across so one side windows installed all the drivers and everything I didn't overnight transfer to move almost two terabytes of games onto the drive so I could do a whole heap of testing so the next day I woke up and I got on with the testing however right away I noticed there was something really bizarre with the results I was getting this thing just we're not even just the results just using this computer itself it just seemed really sluggish loading games took way longer than I was expecting them to since I was using an SSD and when you first got into the game it was very very stuttering I not only got eight gig of ram and sometimes thank you give RAM can cause those problems but even in games like for tonight that aren't super memory hungry it was yeah very laggy for the first 30 to 60 seconds and then every now and then there'd be a bit of a frame started which you kind of expect with an older quad core system but the point was it wasn't running as well as I'd expected it to and then there the performance that I was getting the framerate performance just was nowhere near what I expected in a lot of the titles and yeah it is a has while quad quartz has an all core clock frequency of 3.7 gigahertz so it's by no means a slow processor he's one of the worst examples I came across and also happens to be the first game that I tested with World War Z is quite a CPU demanding title due to all the NPC's running around like mad trying to eat your flesh basically it had the Core i5 4690k at a hundred percent for the entire test though I did see this in quite a few other games tested and some of them I wasn't really expecting to see high C for utilization even with an older Haswell quad core anyway this is a worst case scenario when it comes to gaming we saw a frame time performance jump by 64 percent when going from the OEM PC to our high-end desktop system though with a 1650 you'd honestly get the same numbers with something like a risin 520 400 gr of modern core i3 hell you really shouldn't be that far off with a house wall quad core so right away I knew something was wrong here but to summarize in our test system the 1650 was on average 26 percent faster than the 1050 Ti not an amazing performance uplift but fairly decent all the same now in the OEM PC the 1650 was able to offer just a four percent performance increased and yeah that that doesn't seem right to me rather mess around the OEM PC any longer I decided to take the CPU SSD in memory out and stick it on a z97 motherboard it was late and rather than investigate what was going on I just wanted to eliminate the hardware as potentially defective now doing so still saw a performance hit but it was nothing like what we saw in the OEM PC and it was exactly the kind of performance hit that I was expecting to see from there as well quad core CPU whereas our high-end test system was 64 percent fast in the HP elite desk when comparing the one percent low performance it was just 11 percent fire Master when using the same CPU and memory on a z97 motherboard at this point I knew it had to either be some sort of issue with the software or a power delivery issue with the OEM system and my money was on the ladder so this time after rebuilding the HP system I did a little bit more investigating by opening up Hardware info and I ran a blender workload to stress the CPU I had suspected that it was a CPU throttling issue as the performance hit in World War Z was so extreme meanwhile non CPU intensive titles such as Forza horizon for the margins there weren't that extreme so with hardware info I opened I noticed the CPU is only hitting an all-call frequency of 3.5 gigahertz whereas on the z97 motherboard it was hitting three point seven years and that's a 6% increase but that doesn't really explain how I was seeing a 15% boost to the average framerate and a 47% boost to the one percent low when taking the CPU and memory out of the OEM PC and sticking it in as at 97 motherboard I then notice that the core i5 sustain CPU package power was 28% higher on this at 97 motherboard now this is partly due to the fact that the z97 motherboard increased the v core voltage by 5 percent but I don't think that accounts for all of the power increase rather I believe the OEM system was verum throttling or imposing some kind of power limit what's interesting about the HP system is that the proprietary power supply only has two power connectors coming out of it a 6 pin and a 4 pin and that's it I mean that is absolutely it there is no 20 or 24 pin ATX connector there's no eight pins there's no molex connectors there's not even any solder power cables coming out of the power supply just a 6 pin and a 4 pin power cable then from the motherboard there's another 6 pin power output which connect to the SATA drive so the entire PC the motherboard the CPU the graphics card the storage the whole thing is powered from a 4 pin and a 6 pin power cable and that's it the motherboard itself packs a pathetic looking three-phase vcore verum with no form of cooling so it's completely naked so aware of all this I knew the system had to be heavily Power Limited so I opened up the intel extreme tuning utility but here I found that all the settings met the intel spec the board was limited to a turbo boost power max of 84 watts with a short power max boost of 105 watts for eight seconds however when I ran Cinebench R 20 I found these limits were well out of reach of the OEM PC within just two seconds of hitting the Run button the XTU software detected current throttling at a package TDP of just thirty-eight watts now normally you can adjust the current limit on the Zed 97 motherboard it was set to 100 amps for example yet on the OEM system this option doesn't even exist that means it's hard locked and obviously this has been done to protect the motherboard and power supply in the end we saw a peek package TDP of just 49 watts and again the maximum or call frequency was 3.5 gigahertz in contrast to that the z97 motherboard allowed the core i5 4690k hit 3.7 gigahertz and a package TDP of 58 watts and no limits were imposed so that meant the package TDP was 18% higher than that of the OEM system now what's really interesting despite only a 6% clock speed advantage and a 28% increase in sustained CPU package power the Cinebench r20 CPU score was boosted by 38% so although the CPU is reported to have all cores working at 3.5 gigahertz it's not operating at full capacity and this is why we need to look at actual performance when a processor is limited by either thermal power or current so in short this is pretty scummy HP we're selling businesses a Core i5 4690k Intel's to blame for this rather it's just the OEM builders anyway bashing HP for this poorly designed elite desk 800 g1 isn't really the point of this video the point is to see how the gtx 1650 performs in a popular OAM pc that doesn't have a PCIe power cable so that being the case let's check out a few more games to get a broader picture of how this thing performs with the fastest discrete GPU can get that doesn't require external power we've already seen a worst case scenario with World War Z but here we see a much better showing for the OEM system with Apex legends in the desktop system was just 13% faster on average when comparing the GTX 1650 numbers and more importantly the 1650 still offered a noticeable performance boost over the 1050 TI and a huge performance boost over the rx 560 so easy upgrade is working quite well whether or not it's worth investing $150 u.s. in a graphics card for a computer of this caliber well that's an entirely different story and we'll discuss that later on in the video now these results are pretty typical of what we saw in CPU intensive titles in our test system the 1650 was 26% faster and the 1050 T on average and 56% faster for the 1% lower result however when we moved to the OEM system the 1650 is now just 18% faster for both the average and 1% low 18% certainly nothing to sneeze at but it does really hurt the value proposition of the gtx 1650 the division 2 isn't massively CPU demanding so here the performance drop off isn't too bad that said we're not testing a terribly demanding section of the game we typically use this section of the game for GPU testing as it's easy to replicate for accurate results so the OEM system will likely fall away a bit more in the more demanding sections of the game especially when playing with friends or against other players moving on far cry new dawn had the Core i5 4690k at 100% throughout our testing and as a result we saw a 23 percent performance uplift when going from the OEM system to our high-end test system for the 1650 this meant while the 1650 did offer a rather nice 23 percent performance boost over the 1050 ti on our test system it was just seven percent faster in the HP elite desk so in this instance the 1650 is really no better than a 1050 Ti and only 23 percent faster than our X 560 I have to say the fortnight results are really surprising I expected to see similar margins to what we saw in apex legends that is to say the performance penalty for the OEM rig wouldn't be that extreme however that was not the case in our test system the 1650 was 26 percent faster than 1050 Ti for the average frame rate and 24 percent faster for the one percent load meanwhile in the OEM system it was 18 percent faster for the average but just 6 percent faster for the 1 percent low this meant that overall the gaming spear was very similar and there was really no telling the difference between the two so due to the poor 1% low performance the 1650 didn't really prove to be a worthwhile upgrade in this title now this is a great example of how these two systems will compare in a heavily GPU bound title that doesn't really use the CPU all that much Forza horizon forward play just fine with a dual core so the heavily restricted core i5 4690k except just fine in the HP elite desk for this title testing with just core for tolls a very similar story to that of titles such as battlefield 5 and fortnight here the test system saw these 1650 beat the 1050 Ti by a 31% margin when comparing the average framerate and 30% for the 1% low however in the OEM system the 1650 was just 17% fast and the 1050 TI for the average framerate and just 3% faster for the 1% low so again the game experience were in the 1650 and 1052 I was very similar in the OEM system again we see a similar story with metro Exodus here the OEM system cripples the 1% low performance at the point where the 1650 really isn't an upgrade over the 10 52i moving on although the one percent low performance also takes a big hit and Resident Evil 2 the GTX 1650 did offer noticeably better performance than the GTX 1050 Ti when installed in the HP elite desk ok next up we have Rainbow six siege and please note we only use the single-player portion of the game as there's no real way to test multiplayer accurately so if the multiplayer mode is more demanding than the OEM system will struggle more than what's shown here anyway for the single-player mode the 1% low performance is still pretty weak but as was the case with Resident Evil 2 the 1650 is still able to deliver a noticeably better experience than the 1050 Ti finally we have shuttle of the term Raider and again we see a noticeable downgrade performance for the GTX 1650 when installing and the IM system in our test rig it was 40% faster than 1052 I for the average framerate and 41 percent faster for the 1% low however in the HP elite desk it was 17% faster and average and just 12% faster for the 1% low so that means while the 1650 was an upgrade it was exactly a big upgrade over the 1050 TI before wrapping up the gaming benchmarks I thought it would be worth looking at how the graphics cards stack up in each system using medium quality presets rather than high to ultra so I didn't have a whole lot of time to do that but I thought it'd be interesting a check out so I used fortnight because that's a game where you probably want a few extra frames and then World War Z that threw up some interesting results and the first round of tests so I thought I wonder how that goes when we reduce the quality setting to medium using the medium quality preset and fortnight allowed the gtx 1650 to deliver well over 100 FPS at all times in our test system meanwhile the OEM system performance was still decent however the 1650 is no longer a worthwhile upgrade over the 1050 TI here the 1% low performance is basically identical while the 1650 which is 14% faster when comparing the average framerate previously using the ultra quality settings we saw basically identical performance within the 1050 Ti and the 1650 obviously the medium settings reduced CPU load is here the 1650 was able to pull ahead though it was only 12% faster and average so again not really much of an upgrade here ok so not much really change with the medium quality settings at least the margins didn't change anyway in fact using the medium to low quality settings that really seemed to play into the 1050 TI's favour anyway quite interesting to look at that but I think it's time we check out a 12-game average I think this graph paints a pretty accurate picture of the situation the 1650 was 34% faster than the 1050 T on our test system when comparing the average framerate and 32 percent faster for the 1% low then in the OEM system it was just 24 percent faster on average but more crucially just 12% faster for the 1% low so overall the gaming experience in the OEM system is only slightly improved by upgrading from the 1050 Ti to the 1650 ok so that's all the testing done with the HP elite desk 800 g1 finally got through all of it so now it's just time to wrap this one up though I suspect it isn't going to be a quick and tidy wrap-up I think to conclude this video I probably need to create a few different scenarios so yeah let's do that so the first of these scenarios that I'm going to this one I'm gonna call the hand-me-down so you were given an HP elect desk 800 G one or a similar om system with a proprietary power supply and motherboard so you don't have money to build a new system and you've just got to make do with this this thing so yeah you've basically got to turn this into a gaming rig in the most cost effective and best way you can now remember the system doesn't owe you anything so spending a little bit of money on it isn't too bad not really a big deal well at least within reason that said you don't really want to spend a hundred and fifty dollars us on it to put a graphics card in there at least I wouldn't want to and I have to assume that most of you guys if you've got a an old clunker for free or not that I want to fork out $150 us for a graphics card especially one that doesn't represent particularly good value in a new system if you were to upgrade three or six months down the track or whenever it may be the gtx 1050 i to me seems to be a better fit for this this sort of rig and I certainly wouldn't buy a gtx 1050 TI brand new either but you can get them secondhand for under $100 u.s. the average selling price on ebay at the start of the you and I did my used GPU buying guide that was a hundred and five dollars but I had a quick look just recently and there are quite a few models selling for $80 us and for me that seems like a reasonable investment for this kind of system alternatively a Radeon rx 560 for around $50 that's a real possibility and this is probably the way I'd go of course I'm talking about a secondhand radio rx 560 but I think for a system like this a secondary graphics card probably makes the most sense and that will allow really great performance when you use low to medium sort of quality settings very playable frame rates there and it will just turn this thing into a really good value gaming system frankly I think that's probably the best investment you can make here makes the most sense to me for this scenario spending $150 u.s. that just doesn't seem realistic to me or certainly doesn't seem like a wise investment anyway second scenario you're after a cheap PC and you're considering buying an HP elite desk 800 g1 or a similar om system with a Dietary power supply and motherboard well my advice to you is quite simply don't do that it's just a terrible idea my advice would be do some research and avoid any PC with a custom motherboard and power supply because chances are they're going to be garbage anyway and they're definitely going to be very difficult to upgrade and you've pretty much locked into that so it's just a really bad idea do your research and avoid that like I said in scenario one if you've given this that's a different thing but we've covered that in scenario one scenario two is you're looking for a cheap PC so yes avoid these at all costs it doesn't make sense I paid $350 Australian so apologies to my patreon members for spending that money but I think it's an interesting experiment every three or $50.00 Australian or a hundred and eighty dollars u.s. so if you want to spend that kind of money on this system well frankly you're beyond my help but anyway not to be rude $180 for this one hundred and fifty dollars for a GTX 1650 and you've got yourself a really really terrible value combo I have to say as a side note one of our long-term and very active patreon community members Sal Kagan he put together a system on our discord chat knee posted a link for others to check out I thought that's a really good system it's worth including in this for a bit more information so basically for just shy of 500 dollars like $550 sorry that's before rebates it was about $500 with rebates he created a mid tower build that would absolutely decimate the HP elite desk 800 G one with the gtx 1650 and it was only about $200 more yeah toradol is a bit of money but considering what you're getting I think it's just worth saving up a little bit longer because your money is going so much further it's not even a fair comparison you'll see the specs up on screen I'll try to remember to put a link in the video description for you guys but yeah it's just it's so much better the system he admitted that he was going a bit overboard with the storage capacity on the SSD the power supply was had it been a Headroom in it and he put 16 gigs of RAM which we recommend but you could have saved a bit of money and gone with a beer so there would be possible to make that new system even cheaper but I don't think it it's probably not worth saving the Torah dollars like I said just save up for a bit longer if a $500 you're getting a really good platform there that you can upgrade in the future and make even better now moving on to my third scenario you've managed to buy an OEM system that's not a steaming pile of trash and it won't strangle a GTX 16:50 to death but it also doesn't have a 6-pin pci power connector so does it make sense to spend $150 to put a gtx 1650 in it and enjoy around a 30% performance boost over the gtx 1050 tii that's almost a tough one I would say but again a secondhand 1050 toi can easily be had or quite easily be had for $80 US and that makes the 1650 almost ninety percent more expensive obviously if you're buying any kind of system like this I think it's fair to say you're on a shoestring budget so in this third scenario does it make sense to spend 90% more on something new that's only around 30% faster in terms of performance and that is over a gtx 1050 i not only that but I would say when using the medium quality settings at 1080p the experience is really very much the same in pretty much all games so it's not like you're getting a noticeably better experience with the 1650 especially if you're going to use those medium quality settings which I think most of you would be in a system like this in this scenario we also have to assume for whatever reason you can't watch a quick YouTube tutorial on how to change an ATX power supply it's pretty simple stuff and if you've got the case door off you're pretty well halfway there yeah there's a little bit more to it than that but like I said I'm confident most people could quite easily upgrade one of these things by watching a quick youtube tutorial and the chances are you can probably find a really detailed youtube tutorial for the exact ôm system you're trying to upgrade the power supply and so it's not like you're watching a generic how to upgrade your power supply and then if you're competent enough to pull that off then you can throw in a $35 power supply and chances are you're improving the whole system overall with that higher rated power supply and then you've opened yourself up to a myriad of options so you can you can put an Rx 570 in there you can put an rx 580 in there you can you can put a GTX 1060 you get the idea you can put pretty much anything in there I don't remember in this scenario you've kind of done the impossible and managed to buy an OEM system that's not a steaming pile of junk so that in itself should warrant a PSU upgrade that will enable you to take advantage of a much faster graphics card like the rx 570 now getting back to my horrible HP elite desk 800 g1 it's a bit awkward I know it's sitting right here but sorry mate you you do suck pretty badly so getting back to this thing for the grand total of seven dollars Aussie shipped I was able to order a 24-pin two 6-pin power supply adapter and yeah you would have seen the b-roll but here is that's your that's your replacement for a 24-pin ATX power connector or even a 20 pin power connector in that system so you can get an adapter that will allow you to put a standard ATX power supply in this very system and it cost me seven dollars to get one delivered from I think it's coming from China that would probably make sense but I got it off eBay seven dollars and when that arrives I will be able to hook up a bigger power supply and I'll connect up the rx 517 and see how it goes I mean overall I'm expecting that performance will still suck but for basically the same price as chucking a GTX 16:50 in there we can get a better power supply and a faster graphics card which may or may not give us a few extra frames I mean for titles that aren't CPU intensive like Forza horizon for they will run quite a bit better so perhaps that will warrant them messing about if you get one of these things for free so anyway it seems pretty clear to me no matter which way you slice it at $150 us the geforce gtx 1650 is a bloody pointless product so I stand by what I said in my initial review my initial evaluation of it having looked into this in multiple different ways in much more depth I think that was still a fair point to make and $100 as I said in my day 1 review it would be an absolutely amazing product excellent value you could then push how great the consumption is and how versatile a card is but $150 it doesn't make sense even for the intended purposes which are apparently to breathe new life into OEM systems finally there is a whole other conversation here but we don't really have time for it in this video but I'll just summarize quickly not that long ago there was a discussion started around Intel's TDP and how it's becoming less and less accurate there was concern and quite rightfully so as it turns out that this would cause trouble for OMS who are only build to the Intel spec with quite literally zero Headroom we saw issues with the Intel Core i7 8700 the non K model it throttles quite badly with the included box cooler and there were reports of a few OEM systems that were about 20% slower than the results reported by those building their own systems but as I said a moment ago this really is a whole other conversation but it's still interesting to find that this is far from a new issue even well before Intel started playing fast and loose with their TDP ratings this appears to have been a real issue held the core i5 4690k entire Haswell core i5 range featuring 84 watt TDP yet this part which is actually the flagship locked part that never hit a packaged TDP of even 60 watts so Intel had plenty of headroom back when they were doing their TDP ratings for the has more parts anyway I'm glad we've got two I'm glad I purchased this system I've got to expose just how garbage these lem systems are I mean for an office type PC I suppose it's ok even though you're not getting the full performance of the CPU that you're paid for but that's a whole different kettle of fish that we're not going to get into in this video but I think if you're looking at trying to purchase one of these for $180 u.s. or even $100 u.s. whatever you might be able to get one for I would just I'd give it a wide berth I wouldn't touch it with a 50 foot pole it's just not worth the time and effort and yeah I think most of you will probably agree with me on that one they should just be avoided at all costs anyway I've had enough trashing this thing I think I'm done with this video it's been a long one I don't know how well or how long it ran for I'm very tired but if you watch the whole thing you enjoyed it then well very happy about that and thank you very much for watching all of it make sure you like the video that would be much appreciated for all the time and effort I put into this bloody thing you can subscribe for more content and if you want to become a patreon member to support videos like this then that would be much appreciated you also get access to our discord chat monthly livestream all that good stuff anyway thank you for watching I'm your host Steve and I'll see you again next time hopefully I won't have an oem system