As someone who's only familiar with Valencia on Google Maps, I found the default orientation very confusing. Is it possible to configure it so that north always points upwards?
Likewise! As someone born and raised in Valencia, I found the orientation with the centered position a bit confusing. That said, the layout makes much more sense when focusing on Estación del Norte and Plaza de Toros.
Congrats on the launch! I was so excited when I saw this post.
That is a quite a nice way to explore a city, a middle ground between regular maps (too abstract, no sense of height) and detailed/realistic street view.
Hey I'm from Valencia! This is pretty nice render of the city, thanks for sharing. I liked the mini game as well. Would've loved to see some other angles of some places though!
> A few months ago I was standing on the 13th floor balcony of the Google New York 9th St office staring out at Lower Manhattan. I’d been deep in the weeds of a secret project using Nano Banana and Veo and was thinking deeply about what these new models mean for the future of creativity.
[Proceeds to generate a whole bunch of same-old-shit with LLMs.]
I guess the conclusion is that creativity is dead.
The streetcar tracks fade out in places and I don't see any streetcars. And I can't see my narrow medieval street from this angle. But overall very impressive.
> An image model trained on a hand-picked pixel-art style then repaints it, tile by tile, each tile continuing its painted neighbours so the city stays seamless.
Nice work. The developer of Isometric NYC did something similar - creating a fine-tune of the Qwen Image Edit model. I'd be curious to hear more detail about the image model you used. (the base checkpoint, LoRAs, etc).
My last visit to valencia (as a tourist) was a long time ago (20 yeyrs) but i just had some minutes of fun playing your two tourist games :-) thanks for showing !!
Dense development right up to farmland. Something you will never see in a US city. Not surprising that Valencia is 4x as dense as Atlanta even though Atlanta is 5x more population.
It is incorrect to name this pixelart. Perhaps diffusion-art. I wouldn’t go so far as to call it "slopart"—that would be derogatory.
First, it has very little in common with the process of creation of actual pixelart imagery, as all here on the map is blurred, no pixels are clearly distinguishable.
Second of all, pixelart is something ppl intent-fully create one layer at a time, very often - one pixel a time. Is akin to as embroidery (which pixelart very much resembles), and embroidery is not the same as textile produced by a Jacquard Loom. And the whole point of embroidery is to do it like this in order to achieve the result, not that you cannot do whatever you want with machine textile, and not that we don't wear it daily, because it was programmed and done by machine.
Present map is perhaps the third similar attempt at something... which has its value, but should be regarded the same as perhaps taking Pink Floyd's lyrics and progressions recreated in Suno... or rather training a model to play like Pink Floyd and then call a diffusion product psychedelic rock.
As s.o. who is both super-fan of pixel-art (such as https://x.com/84Aleha/status/2038268570094723313, though I can take isometric stuff also, that is fine), and diffusion imagery (I actually did some poster designs with it, not many), I feel the need to express my disagreement and overall confusion reg if this is an actual pixelart. To me - it is not,.
2. I read the info, but I'm incredibly curious for more detail about the pipeline, in particular how much manual work was done, and can this be generalized to other locations? :)
I guess you're thinking of SimCity 2000? The original used a flat top-down view (with perhaps a bit of "forced perspective" in the artwork for individual zones).
Thanks for sharing! I'm from the region, and I really enjoyed the map and the games! In a year or so, the map will need a stadiums update :D Amunt Valencia!
First thing I did was zoom in on the City of Arts and Sciences and I can tell you that the science museum definitely doesn't have waterfalls along it's sides haha
idbnstra | 5 hours ago
cozzyd | 5 hours ago
akfanta | 5 hours ago
[OP] leugim | 5 hours ago
lausobo | 3 hours ago
OP great work, I love it.
darkwater | 4 hours ago
matallo | 3 hours ago
Congrats on the launch! I was so excited when I saw this post.
mapcars | 5 hours ago
johnny_lol | 5 hours ago
franciscop | 4 hours ago
green-frog-445 | 2 hours ago
harlequinetcie | 4 hours ago
We should figure out a good game to play with it. Like adding transport infrastructure or you need to achieve X time reduction in commuting times.
Nice!
torben-friis | 4 hours ago
jnmandal | 4 hours ago
vunderba | 4 hours ago
https://cannoneyed.com/projects/isometric-nyc
FLeXMurphy | 4 hours ago
[Proceeds to generate a whole bunch of same-old-shit with LLMs.]
I guess the conclusion is that creativity is dead.
jorisboris | 4 hours ago
But this one is particularly detailed, very nice!
vunderba | 4 hours ago
https://news.ycombinator.com/item?id=46721802
https://cannoneyed.com/isometric-nyc
joeadler | 4 hours ago
dn3500 | 4 hours ago
vunderba | 4 hours ago
Nice work. The developer of Isometric NYC did something similar - creating a fine-tune of the Qwen Image Edit model. I'd be curious to hear more detail about the image model you used. (the base checkpoint, LoRAs, etc).
Michael_Houben | 4 hours ago
consensus1 | 3 hours ago
https://tinyvalencia.com/?lat=39.498684&lng=-0.384359&zoom=1...
stagas | 3 hours ago
larodi | 3 hours ago
First, it has very little in common with the process of creation of actual pixelart imagery, as all here on the map is blurred, no pixels are clearly distinguishable.
Second of all, pixelart is something ppl intent-fully create one layer at a time, very often - one pixel a time. Is akin to as embroidery (which pixelart very much resembles), and embroidery is not the same as textile produced by a Jacquard Loom. And the whole point of embroidery is to do it like this in order to achieve the result, not that you cannot do whatever you want with machine textile, and not that we don't wear it daily, because it was programmed and done by machine.
Present map is perhaps the third similar attempt at something... which has its value, but should be regarded the same as perhaps taking Pink Floyd's lyrics and progressions recreated in Suno... or rather training a model to play like Pink Floyd and then call a diffusion product psychedelic rock.
As s.o. who is both super-fan of pixel-art (such as https://x.com/84Aleha/status/2038268570094723313, though I can take isometric stuff also, that is fine), and diffusion imagery (I actually did some poster designs with it, not many), I feel the need to express my disagreement and overall confusion reg if this is an actual pixelart. To me - it is not,.
ericmay | 3 hours ago
adwn | an hour ago
This kind of gate-keeping is getting so tiresome…
nekooooo | 16 minutes ago
nico | 2 hours ago
And make it actually playable? Or let me know if you’d like to collaborate, it’d be a fun project
NikolaNovak | 2 hours ago
2. I read the info, but I'm incredibly curious for more detail about the pipeline, in particular how much manual work was done, and can this be generalized to other locations? :)
zahlman | 2 hours ago
torcete | 2 hours ago
villosil | an hour ago
khazhoux | an hour ago
kubenest | 57 minutes ago
petargyurov | 57 minutes ago
First thing I did was zoom in on the City of Arts and Sciences and I can tell you that the science museum definitely doesn't have waterfalls along it's sides haha
paulbjensen | 19 minutes ago