We’re on tour this week, at my home away from home, and this is my computer away from my computer.
Welcome to Esk Computers, I’m Scott and today we’re seeing if this eight year old HP Omen laptop is still up to snuff, and give it a bit of maintenance while we’re at it.
I’d bought this catchily-named HP OMEN 15-dc0003na on eBay in February 2023 for £180, which I think was a good deal which I rapidly turned into a bad one. It was advertised as the base model, or close to it, which means a 15.6″ screen with a resolution of 1920×1080 and a nice refresh rate of 144 Hz, although that’s a bit hard to make the most out of with its mobile 4GB GTX 1050 Ti graphics. An odd pairing, but who am I to question HP?
That’s pared with a capable Intel Core i7-8750H, with 6 cores and 12 threads, and a base clock speed of 2.2 GHz that boosts up to 4.1 GHz, still a quite capable part today.
It sports a backlit chiclet style keyboard, with the appropriate highlighting of the WASD keys to reinforce that this is for teh leet gamerz. It is functional, I suppose, although to be honest I’ve mostly used this with an external mechanical keyboard or a joypad, if it’s hooked up the TV in the lounge, so in truth I’ve barely used the keyboard for games. And as for typing, well, again it’s functional, but not pleasant. Thankfully I have other solutions for that.
Similar story with the trackpad, it performs its designated function without distinctions or merit, but you’ll probably be using a mouse anyway. I certainly am.
The speakers are very much things that exist. I believe the youth of today would term them “mid”. I’ve heard worse, I’ve heard better. Like a lot of laptop speakers, they are fine until you expect the bass to kick in, where it just doesn’t. Judge for yourself with this ISO standard computer speaker test:
CRAB RAVE
The 720p, 30fps webcam and mic are, well, judge for yourself, as you are watching and listening to them right now. Serviceable is what I’d go with, at a push, at least in decent light.
This was advertised as having a busted fan, so I’d budgeted to replace that although as it happens stripping it apart and cleaning it solved that issue. It was also advertised as coming with the stock 8 GB of DDR4, but on opening it up there was an additional 32GB module present. A nice result, and for a while the most RAM in any system I had.
If I’d simply stopped there, this would have been superb deal in 2023. But I cannot resist tinkering, so I just had to make it a mediocre deal by spending another £25 on a third party battery, which to be fair did give an extra 50% of battery life under full load. That means it now gets an hour and half of flat out use before dying, rather than an hour, and the added bonus of it whining about having a non-HP battery on every boot. Not a great use of funds, to be honest.
I’d also noticed the SD Card reader wasn’t working, so rather than just live with using the many USB card readers I already have I spent a tenner on replacing that, and some of the case screws were missing so I’d spent another ten quid on buying replacements for those, albeit with many dozens of spares for stock.
It also came with the default storage of a 128GB NVMe SSD and a 1TB HDD, although that SSD was just hot garbage, barely any faster than a hard drive, so I swapped it out for a £29 512GB Silicon Power NVMe SSD and the hard drive with a £48 1TB SATA SSD, also from Silicon Power. This latter one failed within a year, so was returned and replaced with a unit that to be fair has worked fine since and now lives in the Kolink Rocket build from many videos ago.
The drive in here now is an ADATA Ultimate SU630 with an awkward capacity of 1.92TB, which cost about £65. Ahh, the beforetimes, when NAND chips were affordable. The same drive’s going for £175 these days, and I’ve decided it’s actually more use to me as an external drive so it’s getting nicked.
So we’re at about £320 as it stands today, and I suppose given the increase in RAM and SSD prices over the last six months it’s actually back to being a pretty good deal, but I’d be a bit more ambivalent about it if you’d asked last year. Sometimes it’s better to stop tinkering and quit while you’re ahead.
However, let’s stretch it out a bit and see what it’s capable of. There’s no surprises to be had, just to be forewarned, although this isn’t cutting edge hardware either now or on its release eight years ago it’s still quite capable.
As such, I’m sure you’ll take my word for it that there’s no issues with office tasks, streaming video, web browsing and all that bread and butter computing stuff.
I’d originally picked this up for photo editing, as the previous laptop no longer ran the photo editing package I use. The Omen runs ON1 Photo quite well, and is only held back in this duty by the screen not being the most colour accurate. The viewing angles however are very decent, no complaints there. It’s working well for this and Affinity, which I use for YouTube thumbnails.
I hadn’t tried video editing on this until now, but I was not wrong in my assumption that it handles DaVinci Resolve without any particular issues. I’m mislaid the footage but from memory my usual 10 minute test project took about 8 minutes to render, three times faster than my HP Envy laptop or 50% faster than my Mac Pro 2013 of previous videos. Very acceptable.
I know that this can play games pretty well, as last year I played through Prey at 1080p outputting to a 4K TV, and it handled it without incident, and I recommend catching up with that game if that’s sitting on your backlog. Here it’s getting about 53 average fps at 1080p High Settings, apologies for the screen tearing, it’s caused by having to go back to the crappier capture device for now. Still, let’s run through the usual tests and see how it stacks up.
Cinebench 2024 gets a still quite useable Single threaded score of 61, even if that’s about half the speed of the current Ryzen 9000 series. The Multi-core result is 358, so while that’s half the score of the slower but many threaded V4 Xeons we have looked at previously, it’s still a solid mix for gaming and general production tasks.
3Dmark’s synthetic tests next, where Fire Strike gets a score of 6915, about three quarters of the result from our previous tests of a 1060 3GB.
It’s a similar story in Time Spy, where it’s score of 2527 is about 69% of the 1060’s score. Nice.
The Superposition benchmark at 1080p High settings gives us 3646, so for the synthetics it’s sitting at about 65 to 70% of the 1060 or the RX580 2048SP, and as expected stomping all over my other HP laptop relying on the Ryzen 2500U’s integrated Vega 8 graphics at 3 to 3.75 times as fast.
The Shadow of the Tomb Raider benchmark at 1080p High gets a reasonable 35 average fps, but obviously settings will need to be cut if you want to take advantage of the high refresh rate screen.
Counter-Strike 2’s FPS Heaven benchmark lets us do that, to a degree at least, getting 103 average fps at 1080p Competitive settings, i.e. Low but with all dynamics shadows enabled
Re-running this at 1080p Very High Settings drops us to a still solid 61 fps, but in this instance I’d be going for frames more than eye candy.
It’s a bit unfair to expect much of a 1050Ti in the Cyberpunk 2077 Benchmark at 1080p Ultra, and indeed we’re only getting 19 fps. This can be improved to 33 fps if we drop to 1080p Low Settings, which is I find borderline acceptable, and that’s a solid showing for a old, low powered mobile part in what’s still one of the more demanding games out there.
There’s normally more of a drop in performance in actual gameplay with Cyberpunk 2077 compared to the benchmark, but oddly that’s not the case here. The usual run round the streets of Kubuki and noising up the rozzers at 1080p Ultra actually gives an extra frame at 20 average fps, but that’s apparently been stolen from the Low settings result which is 32 average fps.
If we are going to be unfair to it, let’s be maximally unfair by throwing Unreal Engine 5 games at it. The Black Myth Wukong benchmark tool at our standard 1080p High, No Framegen, no raytracing settings gets just 7 fps on average.
We can increase that to 44 average fps through monstrous trickery like low settings and 33% FSR, which is probably high enough that we don’t need to resort to framegen. And call me crazy, but this doesn’t actually look too bad despite that massive drop in render resolution.
Marvel Rivals throws a bit of attitude at us, telling us to upgrade the GPU. Very cheeky, but I take the point, at the usual 1080p Medium Settings at Native Resolution we’re down at an unlucky 13 average fps.
Again without resorting to framegen, we can get this up to 30 fps at 1080p Low settings with performance FSR, and if you accept a further hit to resolution, 43 fps at 720p Low Settings with Performance FSR. Although at this point, I think I’d just be guessing what’s going on, it’s a bit of a visual mess.
Forza Horizon 5 at 1080p High Settings gave us 39 average fps, which for fossils like me who lived through the N64 era is fast enough that I didn’t feel the need to optimise further, but I’m sure Low settings will get you closer to 60 fps.
Similar story with Red Dead Redemption 2, although the 27 average fps result at 1080p with the settings slider just into the favour quality side is a smidgen lower than I’d like. Moving down to favour performance gets us 33 average fps, which will just have to do although I expected a bit more of a jump.
Horizon Zero Dawn’s benchmark at 1080p favour quality gives us 33 average fps, but thankfully it moves the needle a bit more at 1080p favour performance settings, giving us 45 average fps.
Baldur’s Gate 3 doesn’t strictly speaking require high fps results to be playable, given the turn-based nature of it, but it’s still better to be faster. At 1080p Ultra setting, taking the usual run around Lower City returns a jerky 19 average fps, however this can be improved to 32 average fps by dropping to 1080p Low. I might actually prefer to take the framerate hit here, this game looks pretty bad at Low settings in my opinion.
The Civilisation 6 Gathering Storm AI Benchmark at 1080p Highest Settings gives us an average turn time of 53 seconds, in line with the earlier CPU test results.
We are rounding things off with the Total War: Warhammer III’s Mirrors of Madness benchmark at 1080p Ultra settings, where we get 27 average fps.
So, in my opinion, this is perfectly cromulent modern, or, well, modern-ish, gaming performance from a not-particularly high end machine even when new, let alone in Space Year 2026.
There’s no dramatic revelation or important moral message to end this video on. Terribly sorry, old bean. It is, however, yet another reminder that you shouldn’t let Big Laptop try and bounce you into an upgrade cycle as long as what you have meets your needs, and this eight year old laptop is still doing a pretty good job of meeting anything I want to do.
I suppose it’s not going to run local AI models particularly well, or at all, but if you think that’s on my to-do list you must be new around here. For nigh-on everything else, there’s not all that much a new several thousand pounds laptop can do for you that this won’t be able to, on some level, and as with all things in life, it’s just a matter of finding your level.
If you have any questions or want further details please leave a comment down below, and if you enjoyed this videotronic missive then consider subscribing. Until next time, take care of yourself, and each other.